I don't think that's quite right. Snow Leopard was a lot of changes to a lot of the OS code base and wasn't great out of the gate, taking multiple dot releases, like all large-scale software updates do, to stabilize and bugfix enough to be "good."
There is no silver bullet, just a lot of lead ones and the answer to Apple's quality problem is to begin baking QA back into the process in a meaningful way after letting it atrophy for the last decade or so.
Hire more humans and rely less on automation. Trust your developers, QA, and user support folks and the feedback they push up the chain of command. Fix bugs as the arise instead of assigning them to "future" or whatever. Don't release features until they're sufficient stable.
This is all basic stuff for a software company, stuff that Apple seems to have forgotten under the leadership of that glorified accountant, Cook.
Well, you can only win playing the stock market (Wall St. is Cook's only real customer) for so long while your products deteriorate. Financializing Apple and eliminating its technical prowess opens the door for the someone else with contemporary technical strength to take Apple's users.
Adding to this, a solution might be enabling continuous releases and leaning into release channels could help in terms of getting more out to users.
In practice it's a challenge because the OS bundles a lot of separate things into releases, namely Safari changes are tied to OS changes which are tied to Apple Pay features which are tied to so on and so on.
It would require a lot of feature flagging and extra complexity which may reduce complexity.
Another way is to start un-bundling releases and fundamentally re-thinking how the dependency graph is structured.
I think they’re painted into a corner with WWDC. Everything has to be a crowd pleasing brain busting wow drop each year. I’m certain there are teams that design their entire workflow around the yearly wwdc. It honestly feels like an executive leadership problem to solve.
If that is a significant part of the problem, then moving WWDC from an in-person keynote attended mostly by nerds and glanced at by the media to an overproduced movie geared at the media and ordinary consumers first probably didn't help. They could've gone back to a stage presentation after COVID, but some of that transition had already been happening prior to that (I recall an increase in how many jokes/bits they were doing in the late 2010's, although that could just be my perception).
Appreciate the sentiment, but in my humble opinion, seems like they should lean into creating even better automated testing, because adding all the new bugs to their suite of automated tests would be a more certain way to decrease their chance of happening again.
But, in a sense, this still incorporates your idea, because the devs and QA must be given the mandate of finding these bugs, and also towards making the automated tests cover the bug's related test cases (as well as charged with improving the test code itself, which is often in a mediocre state in most code bases I've seen at least).
I think some people would be surprised how effective reaching out to apple is for squashing bugs. Three times now I've been assigned an engineer to pin point the bug I was experiencing, after which it was fixed in the next dot release.
By all means people should complain on forums (why not?), but a forum post complaining about some years-old bug isn't going to be anywhere near as effective as contacting apple's support or filing a bug report.
I'm not a developer, I'm just a regular user - so if I can get all this special treatment, so can you.
Interesting. Apple podcasters frequently rant about what a black hole Apple's Radar bug system is. We're talking hours-long rants in some cases. Luck of the draw, maybe? I'm not doubting you, just surprised to read it.
(It feels similar to how those same podcasters absolutely blast Apple Intelligence, while non-tech users I've heard from seem to love it.)
Yes, I am very surprised to hear that you've had such success with reporting bugs to Apple. That is very unlike my experience. I've had exactly one macOS bug that I reported fixed, and that required going to a WWDC lab, talking to a person on the relevant team in person, and having them dig the bug report out of the backlog for a completely unrelated team that it was incorrectly assigned to.
They would be surprised because it's not true, those years-old bugs in the forums have been reported many times to the official bug tracker, with reference number sometimes posted in those very forums.
> the answer to Apple's quality problem is to begin baking QA back into the process in a meaningful way after letting it atrophy for the last decade or so.
As a former Apple employee of 13 years: Apple knows about the bugs. QA isn’t the problem.
A lot of people complain that their radar for some obvious bug isn’t getting noticed, and conclude that Apple must not be QA’ing, or not dogfooding their own product. This isn’t the case at all. I guarantee the bugs you care about are well known, and QA has already spotted them.
The reality is, they just don’t care. The train leaves the station in September. You’re either on it or you’re not. If you spent the year rewriting some subsystem, and it’s July and you have this huge list of bugs, there’s a go/no-go decision, and the answer is nearly always “go” (because no-go would mean reverting a ton of other stuff too, and that carries its own regression risk, etc.)
So instead there’s just an amount of bugginess that’s deemed acceptable. And so the software is released, everybody slaps high-fives, and the remaining bugs are punted to next year, where they will sit forever, because once we do one release with a known bug, it couldn’t be that important, right? After all, we shipped with it! Future/P2, never to be seen again.
An attempt was made to remedy this by pushing deadlines earlier in the cycle, to make room for more QA time, but that just introduced more perverse incentives: people started landing big features in later dot-releases where there’s less scrutiny, and even more tolerance for bugs.
The honest answer is that Apple needs to start giving a damn about the quality of what they’re pushing. As Steve once said at a pretty famous internal meeting, “you should be mad at your teammates for letting each other down like this”. And heads need to roll. I can only hope that they’re realizing this now, but I don’t feel like the culture under Tim works this way. People’s feelings are way too important, and necessary changes don't get made.
Hard to disagree. You would think for a company obsessed with performance per watt and battery life that every release would be as fast if not faster that its predecessor and more efficient to boot.
…and speaking of Snow versions, bring back those cool welcome videos when you first purchase a Mac!
I miss those. Unfortunately, since Apple doesn't do the whole space theme anymore, you'd probably get some really boring drone shots of California at best before a Setup Assistant faded into view from behind a Redwood or something.
> In the 22 years since I became a “switcher”, this is the worst state I can remember Apple’s platforms being in.
Indeed, I remember three times when Apple went a bit overboard on the feature front, but dialed it back and made some of the most stable and useful OS versions:
OS 8.5/8.6 pushed a bunch of features and were the last big pushes pre-OSX, but then OS 9 fixed a TON of bugs, and added a few smaller quality of life improvements that made running 'Classic' Mac OS pretty good, for those who were stuck on it for the transitional years.
Mac OS X 10.0 rewrote _everything_, and especially 10.0 was _dog_ slow, with all the new Quartz graphics stuff in an era where GPU accelerated 3D display widgets wasn't quite prevalent. 10.1 patched in a bunch of missing features (like DVD Player—it was still a pretty useful tool back then), and fixed a couple of the most glaring problems... but 10.4 Tiger was the first OS X release that was 'fast' enough OS X was a joy to use in the same way OS 9 was at the time. At least on newer Macs.
And then of course Snow Leopard, which is the subject of the OP.
macOS 13/14/15 have progressively added more little bugs I track in my https://github.com/geerlingguy/mac-dev-playbook project; anything from little networking bugs to weird preferences that can't be automated, or don't even work at all when you try toggling them.
That's besides the absolute _disaster_ that is modern System Preferences. Until the 'great iOSification' a few years back, Apple's System Preferences and preference pane were actually a pleasure to use, and I could usually remember where to go visually, with a nice search assistant.
Now... it's hit or miss if I can even find a setting :(
Mac OS X versions before Jaguar supported GPU accelerated applications, but the windows were composited in software which caused severe performance problems. Jaguar introduced something called Quartz Extreme, where the windows are treated as OpenGL surfaces and the window contents are textures mapped onto the surfaces. This made OS X significantly smoother on computers with a fast enough GPU and enough VRAM to support it, as the CPU didn't have to spend a bunch of time copying all the window contents to the framebuffer.
Settings is not that bad. It's _awful_, yes, since it broke the panel design we had since the NeXT days, but for me the real annoyance is the way Apple progressively, inexorably broke desktop automation to a point where they now effectively painted themselves into a corner regarding having enough of a foundation to make Apple Intelligence useful (https://taoofmac.com/space/blog/2025/03/14/1830).
That said, I expect things to get worse as they manage to converge their multiple platforms in exactly the wrong way (by dumbing them down across the board even as people keep hoping they'll make iPad OS more useful, etc.).
But at least we still have Safari, Apple Silicon is pretty amazing and I can survive inside Terminal and vim. For now.
There are some factual "gaps" there about how good Snow Leopard was, but I understand the sentiment. As someone who's been a Mac user since System 6 and has been consistently using Macs alongside PCs _daily_ for over 20 years I can say that Apple's software quality (either in terms of polish or just plain QA) has steadily decreased.
It's just that me and other old-time switchers have stopped complaining about it and moved on (taoofmac.com, my blog, was started when I wrote a few very popular switcher guides, and even though I kept using the same domain name I see myself as a UNIX guy, not "just" a Mac user).
For me, Spotlight is no longer (anywhere) near as useful to find files (and sometimes forgets app and shortcut names it found perfectly fine 5 minutes ago), and there is no longer any way to effectively prioritize the results I want (apps, not internet garbage).
Most of the other examples in the article also apply, but to be honest I've been using GNOME in parallel for years now and I consider it to be my "forever desktop" if PC hardware can ever match Apple Silicon (or, most likely, if I want something that is _just a computer_).
Does not work. I happen to know a fair bit about mdutil and the like and confirmed that does exactly nothing for my particular issue. A full Spotlight index reset works temporarily, but after a while it just conks out again.
Also, I vaguely remember there being a way to _order_ results, not just disable them.
I have a vaguely related, kind of interesting story related to search indexes, but on windows instead of mac.
My C drive was was super full for some reason I couldn't understand, and Explorer couldn't tell me where the data was. There was about 100GB just unaccounted for.
Spotlight straight up broke on both of my Macs after Sequoia. It can't even find exact matches in many directories marked for indexing and re-indexing did nothing. Just searching for apps under Applications doesn't seem to find all apps.
I’ve had so many issues with it as well! To the absurd level where I could not search for settings in the Settings app… People all over the net have had all kinds of issues and there’s never been any help other than „oh go and reindex”.
Just yesterday I was trying to find a file in Finder, using the search, and it could not find it even though I was just one directory up from the directory it was sitting in. It made no sense to me at all. Reading these stories, it’s clicking for me.
I gave up on it because of this and installed Raycast which seems a lot more reliable. I used Spotlight effectively as my launcher for apps/settings, and have the Dock completely hidden and Spotlight set to hide everything else. But when it can't even do that consistently, I have no idea how!
iOS has this problem as well. You search for a setting in the Settings app. It’ll say “doesn’t exist” (or whatever) while it’s looking for something extremely obvious (like “software update”) instead of just showing a processing icon.
Then when it does show the results, they’re usually in some terribly unhelpful order. It took me ages to try and go through the CUJ of “this app isn’t sending me notifications because I turned them off now I want them back on”
I've been using Macs since Mac OS 9, and Snow Leopard was indeed very good. It remains my favorite version of Mac OS. I actually think it was Snow Leopard that started the rush of developers to Mac as _the_ platform to use.
People don't want animojis, and they don't want other trite new features that only seem to exist because Apple feels it needs to demo something new every year.
What they want is something that just works without annoyances, distractions, failures, or complications.
Give them that and they'll break down the doors trying to get their hands on it, because it's so far from how most tech works today.
Why would it be bad business for Apple? Their business model is based on selling a holistic ecosystem. They don’t have any need to chase new features and there steady stream of high margin hardware revenue is at stake.
Animojis really feel like peak corporate board asking, "What do the kids like these days?" and dumping that shit into the world. Honestly ... AVERAGE age of the Apple board is 68!! This is a company that's reached some sort of corporate red giant stage where it's influence is massive but it's ability to grow is over and it's only real purpose is to generate heavy metals and seed them throughout the rest of the universe after it's eventual explosive death.
> Most of the other examples in the article also apply, but to be honest I've been using GNOME in parallel for years now and I consider it to be my "forever desktop" if PC hardware can ever match Apple Silicon (or, most likely, if I want something that is _just a computer_).
I'm there as well. I've been really enjoying desktop Linux lately, but I can't go back to a non-Apple laptop at this point. There's just nothing else on the market that comes close, they all make some tradeoff I'm not willing to make - either screen, speakers, keyboard, heat/battery life/fan noise, touchpad, etc. Apple is the only one that has the entire package.
There's Asahi, but no thunderbolt yet and I'm not sure the future of that project with the lead burning out and quitting. I just want an Apple Silicon-esque laptop, no trade offs on components, that runs Linux, and there's no OEM out there that's offering that experience.
So, until that happens I'm staying on mac, and even with declining quality, it's not all that bad compared to the alternatives yet. I've learned to mostly work around/ignore the odd bugs.
I have some hope that Framework and AMD can fix some of those issues. Would love to try out their new desktop (because it's a simpler, more tightly integrated thing) and replace my Mac mini -- then wait for Linux power management to improve.
Linux power management is pretty good. The problem is that defaults favor desktop and server performance. On a MacBook Air 11, my custom Linux setup and Mac OS had the same battery autonomy, despite Safari being much more energy efficient.
The real problem is that, just like the grandparent post pointed out, Apple's software quality has been declining. The Tiger to Snow Leopard epoch was incredible. Apps were simple, skeumorphic, and robust.
Right now, the whole system feels a lot less coherent and robustness has declined. IMHO, there are not so many extra features worth adding. They should focus on making all software robust and secure. Robustness should come from better languages that are safe by construction. Apple can afford to invest on this due to their vertical integration.
I’m pretty sure the touch target only covers the text label. Tap anywhere other than the text labels and it does nothing but close the menu. Really bizarre.
IMHO, Mac OS X contributed decisively towards making Apple cool, which was followed by lots of boutique apps and the success of iOS. Loosing that critical mass of developers, even if it's a tiny userbase, would worry me if I was a top leader of Apple.
Apple has had a contemptuous attitude towards developers since.. the App Store? when the iPhone was out? The last two decades? They don't seem to care about this.
I disagree - if the Vision Pro had some strong use-cases then developers would hold their nose and make apps for it. The platforms that get apps are the ones where businesses see value in delivering for them. Of course businesses prefer it when making apps is easier (read: cheaper) but this is not a primary driver.
I think the potential high-return use-cases for VR and AR are (1) games, (2) telepresence robot control, (3) smart assistants that label (a) people and (b) stuff in front of you.
Unfortunately:
1) AVP is about 10x too pricy for games.
2) It's not clear if it can beat even the cheapest headsets for anything important for telepresence (higher resolution isn't always important, but can be sometimes).
Irregardless, you need the associated telepresence robot, and despite the obvious name, the closest Apple gets to iRobot is if someone bought a vaccum cleaner because Apple doesn't even have the trademark.
3) (a) is creepy, and modern AI assistants are the SOTA for (b) and yet still only "neat" rather than actually achieving the AR vision since at least Microsoft's Hololens, and because AI assistants are free apps on your phone, they can't justify a €4k headset — someone would need a fantastic proprieraty AI breakthrough to justify it.
App Store was a big improvement for developers when it was new, relative to the alternatives.
The things it does may not seem important today, but back then even just my bandwidth costs were a significant percentage of my shareware revenue.
ObjC with manual reference counting wasn't much fun either; while we can blame Apple for choosing ObjC in the first place, they definitely improved things.
This is a ret-con. If you - as a user - were philosophically and inherently against the App Store, then it may seem that way, I guess?
The reality is that there was a long period of time where Apple built up lots of goodwill with a developer ecosystem that exceeded by many orders of magnitude the pre-iPhone OS X indie Mac developer scene.
There were many, many developers that hadn’t even touched a Mac before the iPhone came out, and were happy with Apple, and now are certainly not.
Another way to see it is that people who programmed for Mac OS already had reasons to be annoyed by Apple (e.g. 64bit Carbon). The iPhone let it get new people, who eventually found out why the pre-iPhone scene felt that way.
The actual argument would be people voting with their wallets and moving away from the Apple ecosystem, but this something impossible at least in the USA due to these bullshit "blue bubbles"
For most of the people here they don't. In popular culture and especially among teens and non-technical twenty-somethings there's this absurd "eww green text!" thing. A blue bubble is a status symbol for some reason, even though there's lots of Android phones that cost as much as iPhones.
At this point this is not an argument anymore, it’s just a thought terminating cliche.
Expecting users to change their daily habits in order to marginally improve the operating system of a trillion dollar company feels naive and a bit disrespectful to people who actually use these machines for work.
Even developers… the vast majority of developers ignored Apple for decades (and Apple was also hostile) and it managed to grow despite that.
Might as well ask people to contribute to Gnome or whatever so in the future everyone can go somewhere better. Feels way more feasible.
But the opposite is assuming that Apple has a "responsibility" towards its existing users and has to acknowledge their expectations from them.
A sentiment which famously led Steve Jobs to respond that he doesn't understand this, because "people pay us to make that decision for them" and "If people like our products they will buy them; if they don't, they won't" [0]
So according to Steve Jobs himself, the only Apple-acknowledged way to disagree with Apple is to NOT buy their products, and by extend into the services-world of today it means STOP USING their products.
Now Steve Jobs doesn't officially run this company anymore, but I don't see any indication that this philosophy has changed in any way.
I don't think that's the opposite. The opposite is admitting that people have more than one reason to choose computers, and "voting with your wallet" only works for easily replaceable items, like groceries, clothing, etc.
Most people are not going to migrate to Android, Windows, Linux or whatever else just to make macOS marginally better.
And it's fine: marginal quality improvements of a product are not the "responsibility" of consumers.
I don't think you're taking their argument in good faith. At least my read on what's being said here is that the psyops lock-in effects that Apple uses are too strong.
It's not just "blue bubbles," but "blue bubbles" seems like a good shorthand to me. It's also things like Hand-off, or Universal Control, or getting Messages on both iPhone and Mac seamlessly, or being on the same WiFi network allowing your iPhone/Watch to work as a tv remote for the Apple TV even if you're just visiting a friend. Features that any platform can and does enable, but that do to Apples vertical can work seamlessly out of the box, across all the product lines, while securing network access in the ways most users will want, creating a continuous buy-in loop wherein the more Apple products you buy, the more incentive there is to buy exclusively Apple.
And it's a collective "you." If your entire family uses exclusively Apple products, then you'll be the only person who can't easily use eg the Apple TV in the living room, or the person "messing up" the group chats with "User reacted with Emoji Heart to [3 paragraph text message]," or the one trying to decide between competing network KVM software platforms so that you can use your tablet when your 12-yo can just set their tablet next to their laptop and get a second screen without any setup. Nevermind that these are all social engineering techniques that only exist BECAUSE Apple chose not to play nice with others, they still socially reinforce a deeper commitment to Apple products with each additional Apple product in the ecosystem.
I say this as someone "stuck in the blue bubble" with eyes open about what's going on. I'll keep picking Apple as long as they're a hardware-oriented company, because their incentives are best aligned with mine for the consumer features they are delivering (for now): consumer integration that sells hardware. It's insidious in its own way, but not like "hardware that sells eyeballs" (Google/Meta) or "business integration that sells compliance" (Microsoft).
Apple is addicted to growth. It is as big as it should be, but it acts like an early stage startup always trying to build some new flashy thing to attract the next customer.
It's not Apple, it's capitalism. "Unlimited growth is the ideology of the cancer cell", yet for Apple (or any corporation), it's not good enough to sell 100,000,000 phones. Next year you must sell 105,000,000. And the year after 112,000,000 (not even 110 or your growth is stagnating).
So you get rid of removable batteries so customers have to toss their phones away more often, you gimp other feature, you spend more money on advertising than you did actually developing the product (read this bit several times until it sinks in how crazy it is, yet that's how we are with every major phone, every major movie, etc), and so on.
In 2016 RedLetterMedia did a breakdown of the movies that year, like top and bottom ten grossing movies. They stated that the advertising budget was the same as the production budget, unless they had knowledge of a different number.
I don't doubt that after 2020 the advertising budgets far outstripped the production budgets - multiple times; I am curious if that trend continues now, now that production isn't hamstrung by covid restrictions.
Capitalism works this way because its customers, the investors, want it to work this way, because growth is how you get compound interest. Investors include anyone with an interest bearing bank deposit, a 401k, stocks, bonds, etc.
No growth means it would no longer be possible for an investment to appreciate.
I think of a similar thing when I see people complaining about how companies don't want to pay good wages. When you go shopping do you buy the $10 product or the $5 essentially equivalent alternative? Most people will buy the $5 one. If you do that, you're putting downward pressure on wages.
It's in your (purely economic) best interest for your wages to be high but everyone else's to be low. That's because when you're a worker you are a seller of labor, while when you're a customer you are an (indirect) buyer of labor.
Everything in economics is like this. Everything is a paradox. Everything is a feedback loop. Every transaction has two parties, and in some cases you are both parties depending on what "hat" you are wearing at the moment.
Growth isn’t necessary for high returns on equity. And it isn’t necessary for the investment to provide a return.
Equity returns ultimately come from risk premiums. (Which are small now in US equities BTW).
I’m invested in a microcap private equity fund that has returned >20-25% for years. They have high returns because they buy firms at 3-4x cashflow. You will get the high returns even with no growth. And with no increase in valuation. The returns are a function of an illiquidity premium.
With Apple explicitly, growth is expected given the valuation level. If it doesn’t grow, the share price will decline. So yes, in their case, firm is certainly under pressure to grow.
I also don’t agree with your “best interest for wages to be high and everyone else’s lower”. That is one aspect. It is more complicated. Consider Baumol Effect for starters.
I'm talking about macroeconomics, not micro. Risk premium means there is risk; not everyone gets a return at all. The entire society, as a whole, cannot experience consistent returns unless there is macroeconomic growth. If the pie is not getting bigger, someone has to be losing for someone else to gain.
Things like retirement, 401ks, etc., are society-wide institutions subject to macroeconomic rules.
I'm sure everyone has seen this 100 times already but it really fits given modern advertising practice of every major company, especially in designing products to fit advertising plans.
There are also entire "industries" designed to shield people who want to find quality content from big 'A' advertising.
I love how he uses the word “craftsmanship”, something that he understood quite well (considering how close he was working with people like Bill Atkinson, Andy Hertzfeld, Burrell Smith, etc).
Today engineers have to put up a fight to do anything resembling craftsmanship.
> What’s the actual argument that will credibly convince the top leaders of Apple, to push fixing MacOS up the list of priorities?
That their own products depend on it because they developer their products in Mac. And that the professional people they pretend they cater to depend on Macs, and steadily move away.
> What’s the actual argument that will credibly convince the top leaders of Apple, to push fixing MacOS up the list of priorities?
Unrelenting bad press. People talking about nothing else but the decline of their software quality. We can already see that with the recent debacle which caused executive shuffling at the top of the company.
That shuffling was caused by Apple utterly failing to deliver a major feature, that was a key selling point for the latest generation of their hardware.
"Bad press" for their declining software quality is like people complaining there's no iPhone mini/SE anymore. Apple just doesn't give a fuck. They've joined the rest of the flock at chasing fads and quarterly bottom lines.
What was the major feature? The complete uselessness of “AI” on macOS? I updated and enabled all the AI features and I would ask Siri from my M1 and it failed every time. Would just continuously try with its annoying ping sound and never work. Blew my mind that they let this out.
Yeah I was talking about the "AI". It's such an utter failure that even Gruber has been calling it out.
It was already the same story with AirPower (the wireless charging mat). They've pre-announced it, even tried to upsell it by advertising it on the AirPods packaging. It just turned out physics is ruthless.
TBH I've been increasingly sceptical about voice assistants in the "pre-AI" era. I sold my HomePods and unsubscribed from Apple Music because Siri couldn't even find things in my library.
> I sold my HomePods and unsubscribed from Apple Music because Siri couldn't even find things in my library.
I have almost the opposite problem this year. I tell the HomePod to turn the office lights on, it sometimes interprets this as a request to play music even though my library is actually empty, and the response is therefore to tell me that rather than turn on the lights.
Back in the pandemic, same problem with Alexa. Except it was in the kichen, so it said (the German equivalent of) "I can't find 'Kitchen' in your Spotify playlist" even though we didn't even have Spotify.
A few months ago, for quite a few years, Siri (in the car) would respond correctly to "Play playlist <playlist name>". Now it interprets that as of about two months ago that it should play some songs of the genre (I have a playlist named "modern").
I think the best argument is to remind Apple that they aren't selling the OS anymore, so they don't need a new version every year. And that macOS features is not what is pushing Mac sales. People aren't buying the M series machines because of the new macOS version, they are buying it because of the hardware. The M series chips are impressive and provide some great benefits that you can't get elsewhere.
And that hardware needs to be coupled with solid software to hook and keep people on this computer. So they can take more time to create more compelling upgrades and sand off more edges.
I think they need to desync all their OS's and focus on providing better releases. There really is no benefit to spending the day updating your Mac, phone, tablet, appletv, and HomePod. Especially when there are no good reasons to update. I feel like Apple became far to addicting to habit and routine that it's become more important to keep that than deliver product. Apple Intelligence is a good example of that.
The iron law of bureaucracy happens because humans have a finite amount of time to spend doing things. Those dedicated to bureaucratic politics spend their time doing that, so they excel at that, while those dedicated to doing the work have no time for bureaucratic politics.
It's related to why companies with great marketing and fund raising but mediocre or off-the-shelf technology often win over companies with deeper and better tech that's really innovative. Innovation and polishing takes work that subtracts from the time available for fund raising and marketing.
Great insight—thanks for sharing. It strikes me that bureaucracy is inherently self-perpetuating- once established, it rewards compliance over creativity, steadily shifting the culture until innovation becomes the exception rather than the rule.
Perhaps the real challenge isn't balancing innovation and marketing—it's creating a culture that genuinely rewards bold ideas and meaningful risk-taking.
It seems most organizations naturally become more risk-averse as they age and grow since the business becomes more well-defined over time and there is more to lose from risky ventures. The culture has to reward meaningful risk-taking even when that risk-taking results in a loss, which can cause issues when people see the guy who lost a bunch of money getting a bonus for trying (not to mention the perverse incentives it may create).
> [Bureaucracy] rewards compliance over creativity
Imho, this is the wrong takeaway from parent's point.
Bureaucracy rewards many things that are actual work and take time. (Networking, politicking, min/max'ing OKRs)
Creativity and innovation are rarely part of the list, because by definition they're less tangible and riskier.
A couple effective methods I've seen to fight the overall trend are (a) instill a culture where people succeed but processes fail (if a risky bet fails then the process goes under the spotlight, not the person) and (b) tie rewards to results that are less min/maxable (10x vs +5%).
Desktops are in S3 half the day consuming ~0 power. During use, electricity costs are so much lower than hardware costs that approximately nobody cares about or even measures the former. Servers have background tasks running at idle priority all day so the power consumption is effectively constant. Laptop and phone are the only platforms where the concept of "Linux power management" makes any sense.
My Mac mini (M1) sips ~6W idle and is completely inaudible. It acts as a desktop whenever I need it to, and as a server 24/7. I only power up my NAS (WoL) for daily backups. The rest of the homelab is for fun and experiments, but mostly gone.
"Idle" x86-64 SOHO servers still eat ~30W with carefully selected parts and when correctly tuned, >60W if you just put together random junk. "Cloud" works because of economies of scale. If there's a future where people own their stuff, minimising power draw is a key step.
Does the mini PC go from zero to eleven though? Can I play BG3, Factorio, or Minecraft on the same hardware? Can I saturate a TB3 port? Transcode video? Run an LLM or text2img? Any of that while remaining responsive, having a video call?
If I already need a powerful machine for a desktop, why would I need a second one just so it can stay up 24/7 to run Miniflux or Syncthing? Less is more.
> Robustness should come from better languages that are safe by construction.
Nahh, robustness comes from the time you can spend refining the product not from some magic property of a language. That can help but just a bit. There was no Swift in Snow Leopard. Nor there is not much Rust in Linux (often none) and even less (none) in one of the most stable OS available, FreeBSD.
They should just release a new version when the product is ready and not when the marketing says to release it.
I want Mac hardware but Linux software. The other makers build quality is horrendous. Especially in the 13inch segment which is my favorite.
Using a pretty old laptop because there is no replacement right now.
The new Ryzen AI looks really interesting!
Sadly there is no Framework shop for me to look at it and they not ship to Japan..
Yeah, I heard good things about it. I do a lot of gamey development stuff and x64 makes that easier. But Asahi seems to be catching up a lot recently, maybe I should look at it again! https://news.ycombinator.com/item?id=41799068
Asahi is an adventure. I am in the same camp where I got a MacBook for the hardware, but am really a Linux guy. I got really excited when the fex/muvm patches came out for Asahi, and switched to mainly booting it for a couple months. 80% of what I needed to do worked, but that 20% still wasn't there. It was mainly the little things too:
1. Display output from USB-C didn't work
2. Couldn't run Zotero
3. Couldn't compile Java bioinformatics tools
4. Container architecture mismatches led to catastrophic and hard-to-diagnose bugs
There were things that worked better, too (better task management apps, and working gamepad support come to mind). Overall, even though I only needed those things once or twice a week, the blockers added up and I erased my Asahi partition in the end.
I really appreciate the strides the Asahi project has made (no really, it's tremendous!), and while I would love to say that Linux lets me be most productive, features like Rosetta2 are really integrated that much better into MacOS so that I can't help but feel that Asahi is getting the worst of both worlds right now. I'll probably try again this summer and see what has developed.
The biggest issue Framework have right now is shipping. I can order a ThinkPad practically anywhere. No so with Framework - they are literally leaving money on the table from what I would assume their core segment: affluent tech savvy users trying to get off the planned obsolescence cycle.
Tell me you're from the US without telling me you're from the US.
Jokes aside, I had to wait years for Framework to finally allow shipping via a friend in Berlin. I think they ship to Sweden now—they seemed to have an unfortunate misunderstanding that they needed to produce a Swedish keyboard and translate their website before shipping here, which of course is poppycocks.
I am pretty sure that if you have reached the point that you are ordering a laptop online from a brand unknown to the general public, it means you are past the point you need the actual physical keys to match your keyboard layout on your OS settings. You could just have blank keys.
To be fair, some international keyboard layouts actually have variations of key shapes and locations. The shape of the Enter key and the cluster around it is the main example. So it's more than just the labels.
I own both ISO and ANSI keyboards on different laptops and use the same software keymap. I don't think it is such an important factor as I switch from one to another without thinking about it.
Out of curiosity, what are you basing this on? From having spoken to people who manage IT fleets, and being the person regular people ask for advice for what device to get, with the occasional exception (which Apple also had plenty of, cf. the butterfly keyboard), you get what you pay for. A 1k-1.5k+ Asus/Dell/HP/Lenovo will get you decent and good build quality.
At work, our Windows devs use expensive XPSs that are complete crap failing constantly, both hardware and software. As someone who used Latitudes and Precisions when these were the reliable workhorses you seem to describe, the new stuff is just outrageous. (My personal laptop is still an e6440).
My work machine is an M2 Pro MBP and except the shitty input HW (compared to the golden era of Thinkpads/Latitudes without chiclet keyboards) and MacOS being quite bad compared to Linux, it completely trounces the neighbouring Dells that constantly need repairs (mostly the USB-C ports and wireless cards failing).
Maybe if you run a fleet that's statistically true. If you're a regular person you can have incredible bad luck with specific models.
Got two "2k" Lenovos at 4 year intervals.
The first one worked fine but that model was known to have a weak hinge. Had to replace it three times.
The second one had a known problem that some units simply stop working with the internal display and the only solution is replacing the motherboard. My unit worked about a week for me. Seller refunded me instead of repairing because it was end of the line and they didn't have replacements.
Got a "2k" Asus ordered now, let's see how that goes :)
Compared to that, even the one emoji keyboard macbook pro that i had worked for years. The keyboard on those models is defective by design and kept degrading, and I still think Cook should take his dried frog pills more regularly, but the rest of the laptop is still working. Not to mention my other, older apple laptops that are still just fine(tm), just obsolete.
I think price isn't the only thing. PC gaming/consumer laptops lean pretty heavily on price to performance ratios and I think they cut build quality to do it. Business lines like Thinkpad/EliteBook tend to offer worse performance dollar for dollar but they are built better.
> A 1k-1.5k+ Asus/Dell/HP/Lenovo will get you decent and good build quality.
And it still won't be on par with a $999 apple silicon air, or a MBP.
I've deployed latitudes, precisions, and thinkpads. They all still make tradeoffs that you don't have to deal with on the mac.
The X1 carbon is probably the "best" but, even with that - you are still getting a 1920x1200 screen unless you spend more than a MBP for the 2.8k display (which is still less than the 14" MBP, and costs more than an equivalent specced M4 pro). The trackpad is worse, the speakers are worse, battery life is worse, and they're loud under load.
They're all fine for a fleet where the end user isn't the purchaser, which is why they exist, but for an individual that doesn't want tradeoffs (outside of the tradeoff of having to use macOS), there's no other option on the market that comes remotely close to the mac. For someone that wants Apple silicon MBP level hardware but wants to run Linux, there are zero options.
The screen is the most egregious tradeoff though, the PC world is still adverse to HiDPI displays and even on high end models 1080p or 1200p is still the standard. I can excuse poor speakers, it is a laptop after all, if I really had to I can deal with fan noise, but I shouldn't have to spend more than a MBP to get a decent 120hz HiDPI screen with sufficient brightness and color accuracy.
Thinkpad line from Lenovo. Amazing build quality, and you can order them with Linux.
I have a P1 Gen 7 and it’s fantastic. It feels premium, and it’s thin, light, powerful, has good connectivity and 4K OLED touch screen. I’d take it over Mac hardware any day.
Non-Thinkpad Lenovos have some standouts too. I'm running Debian Stable on an AMD Yoga Slim 7 from a couple of years ago and sure, it's not an Apple, but for the £800 or so I paid for it, it's a really polished machine. Loads of ports, and it's approximately performance-competitive with a Dell XPS13 from about the same time that cost literally twice as much.
The one snag I ran into was that when it was new, supporting the power modes properly needed a mainline kernel rather than the distro default. But in the grand scheme of things that's relatively trivial.
I have an M1 Macbook Pro from work and honestly I'm not tempted to get one for myself. I am tempted by the M3 and M4 beasts as AI machines, but as form factors go I'm just not sold.
Aren't the only Thinkpads with displays in the 4k neighborhood 16-inches? The 14-inch Macbooks are 3024*1964 and have all been like that for a while. I don't know why the PC world (and Linux ready by extension) undervalues high DPI so much, because it makes it hard to consider going back.
The screen keeps me on macbooks as well (well, and the touchpad, the speakers, and the lack of fan noise).
But it is baffling how 1920x1080 (or 1200p) are still the "standard" elsewhere. If I want an X1 carbon, the best screen you can get at 14" right now is 2880x1800 (2.8k). Spec it with 32GB of RAM and it's clocking in at $2700, for a laptop that still has a worse trackpad, worse sound, and worse screen than a 14" MBP at $2399. And the Ultra7 in the thinkpad still doesn't beat the Mac, and it'll be loud with worse battery life.
There truly is nothing else out there with the same experience as an Apple Silicon MBP or Air.
So, my only options for the foreseeable future is wait for Asahi Linux, or suck it up and deal with macOS because at this rate I don't think there will ever be a laptop with the same quality (across all components) of the mac that can run Linux. The only one that came remotely close is the Surface Laptop 7 with the Snapdragon elite, but no Linux on that.
Consider a thinkpad or lenovog yoga pro. I don't think the difference is is that pronounced anymore, maybe it never was, but you always need to look at the premium segment. Somehow people end up comparing budget pc laptops and macbooks.
What dou you mean with more integrated? It is a regular desktop PC with an apu (like is totally common for office PCs, just bigger) and soldered instead of upgradeable ram.
It would be kind of funny, but also very sad, if Apple guys mistook the copying of apple's worst behaviour - producing throwaway devices - as a sign of quality. Though I think we are there for years now with phones, I wouldn't expect such thinking here.
It is "integrated" in the way that the processor is an APU that has specific memory bus requirements. That's all. It is not an integrated software-hardware system that is finetuned, and that board is not any better than a a generic motherboard would be for a regular processor.
My point is that this system is not integrated in the way apple fans usually define the word. I'd claim it is not integrated at all. It is a regular PC (but with soldered ram), which is exactly like framework announced it.
There should be no need to sprinkle some apple marketing bs on that to make it attractive.
I really wish everyone would stop entertaining these borderline crackpot hypotheticals that all rely on the notion of “those damn Apple dummies not getting it!”
As someone who actually studied human computer interaction, and since I had to work with borderline unuseable macs multiple times in my career now, plus as someone seeing the utter failure of relatives in just using an iPhone (bought since "it is so much easier", now not even able to call from the car system since it is so buggy), the Apple popularity is absolutely a case where you have to look at external factors like social status. And if that translates to "the users are dummies" to you, then that's your interpretation. Plus yes, translating marketing/status concepts like a bogus "integrated" status absolutely is interesting, thus my intent to clarify whether that is really happening here (plus some criticism, admittedly).
Probably not worth it going further into this though, it will only derail.
This really is exactly how I feel. There are too many tradeoffs to switch to non-Apple hardware at this point. I'd love to run Linux/BSD full-time, as many of the apps that I frequently use on my Mac are FOSS (e.g., R, PyCharm, darktable, etc.) I've been a Mac user since 2002, and Mac OS X served as my gateway to the Linux/BSD world (that, and a short-lived use of RH 6.2 on an old Dell laptop). IMO, macOS really does need a Tiger/Snow Leopard-esque release, but I'm not sure the vast majority of macOS users would even appreciate such a release.
Also while Apples software quality has definitely diminished over the years, Windows in the same period has utterly CRATERED. Like I get along fine with 11 for my gaming PC but with every single update one feature or another becomes notably broken.
Fedora atomic distributions in general are great. I recommend Bluefin-dx over Bazzite (they’re both GNOME-based Fedora Atomic from the same group— universal blue) for developers, because it’s really easy to install the packages that Bazzite gives you, and it comes pre-installed with Docker.
Is it comparable to gaming on Windows? Last time I tried the performance wasn't as good for some games (Deadlock) and it took ages to compile shader (it takes 30 seconds on Windows with the same specs)
I saw long shader compile times for at least one game last month, might have been Deadlock. I have a Radeon RX 7600 & Ryzen 9 7900X3D for reference.
There is mention on the arch wiki about enabling multi-threaded compiles, but also I have read you perhaps dont even need to precompile them now and possibly get better performance as the JIT compiles via a different vulkan framework (VK_EXT_graphics_pipeline_library).
I disabled pre-caching (which effects the compile too afaict) and never noticed any stuttering, possibly past some level of compute it's inconsequential. I also noticed that sometimes the launcher would say "pre-compiling" but actually be downloading the cache which was a lot slower on my internet.
Certainly on my (very) old intel system with a GTX1060, Sekiro would try to recompile shaders every launch, pegging my system at 99% and running for an hour. I just started skipping it and never really felt it was an issue, Sekiro still ran fine.
Yep Bazzite is great. But the difference between them is mostly just the packages installed. To me it’s easier to install the gaming related packages from Bazzite onto Bluefin.
I have a problem with Docker sockets while installing onto Bazzite, and didn’t care enough to look further into it.
Windows is indeed an execrable shitshow. Every aspect of it assaults the user with incompetence or outright hostility.
First is the endless badgering to log in, LOG IN, LOGGGG INNNNN with an asinine Microsoft account. If you can tolerate that and actually get the OS running, you're wading through a wonderland of UI regressions and defects.
The default hiding and disabling of options is infuriating. Try showing content from your Windows computer on a TV, for example. You plug your HDMI cable in, and you can select the TV as an external monitor in a reasonably logical manner. Great.
But wait... the sound is still coming from the laptop speakers. So you go to Sound in the system settings. Click on the drop-down for available devices. NOPE; the only device is the laptop speakers.
So you start hunting through "advanced settings" or some such BS. And buried in there you find the TV, detected all along, but DISABLED BY DEFAULT. WHY??? Not auto-selecting it for output is one thing, but why is it DISABLED and HIDDEN?
This is the kind of shit I have to talk my parents through over the phone so they can watch their PBS subscription on their TV. The sheer stupidity of today's Windows UI isn't just annoying, but it's demoralizing to everyday people who blame THEMSELVES for not being "computer-savvy" or slow learners. NO; it's Microsoft's monumental design incompetence and user-hostile behavior.
Microsoft doesn't get the relentless excoriation it deserves for its miserable user experience. There's no excuse for it.
I can't say I've ever had HDMI audio mystery-disabled when I try and use it, that's for sure a new one for me. That said the entire audio stack is an utter fucking nightmare. Selecting sound devices usually works, unless the game/software you're using either isn't set up to know about it, or isn't being told by Windows, either or. Then of course there's the fun game you play having two HDMI displays where Windows will constantly re-enable one you've disabled because it doesn't have any fucking speakers on it.
Win 11 must have some sort of contextual HDMI audio switching where it figures out exactly where you want your audio to go and then does the opposite. Because my Win 11 work laptop loves to re-enable HDMI audio and make it the active audio connection despite the fact that neither of my monitors have built in speakers.
Might be, still I no longer feel like baby sitting Linux on laptops as I used to, yes I had another go at it just last year I know how well it works, and I will never pay the Apple tax outside assigned work laptops.
My job gave me an expensive high-specced laptop with Windows on it. This is the first time I am stuck using Windows daily. It's W10. With Windows Defender and a bunch of windows, it starts to slowly become unusable. Today, it blue screened for me just fixing again (and again and again) the bluetooth headphones never gets automatically switched to when I turn them on. Forget about having Visual Studio open on it for an extended period of time.
Meanwhile, my 7-year old laptop with Fedora on it I type this is wonderfully snappy and stable. I started to get tempted to actually switch back to a Mac just to get some predictability and stability, but I have avoided macs for years. (And - never having to deal with constant line ending issues)
All I hear from other co-workers is how their perfectly specced laptops lag with Windows. It's freaking Stockholm Syndrome here!
oof I re-read this comment. I would only switch to the mac because the company didn't setup any of the required security rigamarole to run on their internal network, but otherwise, nah. I dropped that years ago - expensive, disposable machines to run emacs, a terminal and a browser ...
Your loss. I haven't been able to tolerate the MacOS experience since Catalina, running GNOME with a Magic Trackpad has felt head-and-shoulders better for the past 3 years at least. Apple Silicon is neat but was never an option for native development in my workflows anyways. The software matters more to me, and MacOS has been sliding down the subscription slopware slope for years now.
I am perfectly happy to use last-gen hardware from Ebay if it runs an OS that isn't begging me to pay for subscriptions and "developer fees" annually. My dignity as a human is well worth it.
As a former Mac user, I'm really happy with my System76 linux laptops. The only tradeoff is the terrible built-in speakers. My Lemur is lighter and has better battery life than my Macbook Air and has been bulletproof despite my ill treatment. Each of my Macs, however, have had various hardware failures or the famous keyboard recall on the horrible touchbar Macbook Pro. I also prefer matte screens to glossy, so that's a win for me, but ymmv.
* Battery life is a lie, especially since it drains almost as much battery closed as it does open.
...
Overall, I think I am probably going to switch back to a macbook after this, not being able to go a day without charging and your laptop always being on low battery is a bit anxiety inducing.
My ideal laptop would be the macbook trackpad, monitor and battery life stuck inside any thinkpad. Or just anything non MacOS, even Windows, in the macbook. I despise MacOS with every fiber of my being, but the hardware is damned good.
The reason that keeps me on Windows, is that you left out of your list gaming and 3D graphics on laptops.
Metal isn't really on pair with Vulkan and DirectX in terms of relevance for graphics programming, the M chips aren't up to NVidia ecosystem, SYCL, the two major compute APIs for any kind of relevant GPGPU workloads, and thus don't really matter.
And gaming, well, even though all major engines support Metal, there is a reason DirectX porting kit is now a thing.
So why pay more for a lesser experience, and then there is the whole issue macOS doesn't have native support for containers, like Windows does (their own ones), and WSL is better integrated and easier to use than Virtualization Framework.
Gaming is pretty great on Linux now. I just finished a little Elden ring session and it still blows my mind that when I close the game my Linux desktop is there behind it. No more dual booting, hopefully will never need windows for anything ever again.
You mean translating Windows and DirectX APIs is great, there is hardly a Linux gaming ecosystem.
Proton is the acknowledgment of Valve's failure to entice game studios, already targeting Vulkan/OpenGL ES/OpenSL on Android NDK, Switch (which has OpenGL 4.6/Vulkan support), or on PlayStation (Orbis OS being a FreeBSD fork) to target GNU/Linux.
Strange take. Proton is an acknowledgment that the windows apis are the de facto for gaming. Not sure why the runtime matters. Some games ever run better. Not sure why that’s not the ”real deal” but whatever I’m glad you’re happy with your spyware gaming OS.
If you're playing the likes of Fromsoft/Resident Evil/Kojima games on a PC, be it Windows or Linux, you're not playing on the platform those games were designed for.
The problem with your reasoning is that Windows/PC doesn't need to emulate Orbit OS and LibGNM, Sony also supports Direct X and Win32 directly on their engines.
"supports" as in I see articles in the PC gaming press about technical problems with From/Kojima games a year after I've finished said games on console with zero issues.
There’s no such thing as “native”, all the things you’re talking about are translation layers for hardware instructions themselves, and the overhead for doing software based translation is significantly less than hardware accelerated virtual machines- and we as an industry love those.
The reason for this is because the translations are very cache friendly and happen in userland, so the performance impact is negligible, and the scheduler on Windows is so poor compared to Linux ones that it’s even common for games to perform better on Linux than on Windows.. Which is crazy when you consider the difference in quality of the GPU drivers.
I understand that you want it to “just work”, but that tends to be the experience anyway.
You can do what you want, it’s your life, but this is not a terribly good excuse. Valves “failure” is essentially rectified.
I will add though, that it’s actually Stadia that made linux gaming the most feasible, many game engines (all of the ones I worked on) were ported to Linux to run Stadia, those ports changed essential elements of the engine that would have been slow or difficult to translate; so when Proton came around quite a lot of heavy lifting had gone away. I only say this because Valve gets some credit for a lot of work our Engine programmers did to make Linux viable.
So how many of those ported game engines are actually making a different on GNU/Linux gaming today?
There is certainly such thing as native, one thing is the platform where the APIs were originally designed for and battled tested, and the other is other platform emulating / translating them, by reverse engineering their behaviours with various degrees of success.
Valve's luck is that so far Microsoft/XBox Gaming has decided to close an eye on Proton, and it will run out when Microsoft decides it has gone long enough.
Unreal can target Linux, sure, but not all of the plugins you might use will, nor any of your own plugins.
Unreal is almost worse because their first party tools (UGS, Horde) will not work on Linux, so you have to treat linux as a console, and honestly the market share isn't there to justify it.
Specifically, I worked on those games so I know what they natively support and how things transpired behind the scenes.
Proton has absolutely no hope of working without the changes we made because of stadia, the code we wrote was deeply hooked into Windows and we made more generic variants of many things.
The Division 1 PS4 release was significantly shimmed underneath compared to the win32 and xbox releases: this became much less true over time as porting the renderer to linux (specifically debian) made us genericise issues across the OS’s and when Div2 shipped we had a lot more in common across the releases; we didn’t rely on deep hooks into Microsoft APIs as much
> this became much less true over time as porting the renderer to linux (specifically debian)
Strange how you ported the renderer to Debian, and yet you couldn't even find a link to a game that has a native Linux support.
Was there ever a port?
> Proton has absolutely no hope of working without the changes
You keep saying this as the absolute truth, and yet at the time when Stadia launched Proton already had 5k working games under its belt.
Strange how Stadia is this monumental achievement without which Linux gaming wouldn't happen according to you.... and yet no one ever mentions Stadia ever contributing any code to any of the constituent parts of what makes Proton tick. Apart from the changes that engines supposedly made to work on a yet another game streaming platform.
I don’t know how to say this without being unkind.
There is a functioning version of The Division 1, Division 2, Avatar and Star Wars outlaws that run on Linux internally at Ubisoft.
Nobody will release it because it can’t be reasonably QA’d. (Stadia was also very hard to QA, but possible, as it was a stable target and development was essentially funded).
I’m not sure what your problem is; I said - as clearly as I can - that architectural changes to the engine were neccessary for proton.
I know this, for an absolute fact, because Proton was a topic when I worked on those games and it was not until Stadia (codename Yeti) was on the roadmap, and our rendering architect lost all his hair working on it - that Proton started to even function slightly.
I’m not shilling for Stadia - there’s nothing to shill for, it is dead.
Get over yourself, if you don’t like the truth then don’t start going in on me because my reality does not match your fantasy. Sometimes corporations do things accidentally that push other things forward unintentionally.
I just want to share my thanks to Stadia because I know for a concrete fucking fact that at least some of the AAA games I worked on would not function at all on Linux without Stadias commercial interference.
All I'm saying is that "it’s actually Stadia that made linux gaming the most feasible" statement is at best contentious because in reality gaming on Linux was already made (more) feasible when Stadia had only just launched.
And Stadia used the same tech without ever giving back to Proton at all (atl least nothing I can quickly discover). So the absolute vast majority of work on Proton was done by Valve which you dismissed as "when Proton came around" (it came around before Stadia) and "quite a lot of heavy lifting had gone away" (Valve did most of the heavy lifting).
That's the extent of my "problem".
> at least some of the AAA games I worked on would not function at all on Linux without Stadias commercial interference.
So, not "actually Stadia that made gaming feasible on Linux" but "because Stadia used all the same tech, and there were possible commercial incentives early on until Google completely dropped the ball, bigger studios also invested in compatibility with the tech stack"
Stadia did a lot to help by being a stable target and by being seen as commercially viable. Google also helped a lot to aid developers, not just financially.
That they didn’t contribute code to proton doesn’t factor at all, I just hate to see people not get their dues for their part in the prolification of Linux gaming- because I saw it first hand.
You are labouring under the delusion that I’ve implied Proton did nothing, no, they levied a lot of existing technology and put in a lot of polish. They were helped by Stadia, by Wine, by DXVK and others.
They didn’t do it alone, that doesn’t minimise their contribution, it contextualises them.
Also: Stadia ports of games were native, they did not use proton- it was architecture changes of the games themselves that made proton work better- not Google making proton itself function better.
That proton was running some games is a weird revisionist take, very few AAA games ran at all, those that did were super old and there was always some crazy weird bugs- proton got better but also AAA games coalesced into conforming to linux-y paradigms underneath better- so support got better much quicker than expected. You can even see this if you track the “gold” released games over years, some of the worst supported games for Proton are from 2015-16; before stadia but after game complexity started rocketing up with next game engines of the day.
Hope that helps, because honestly this conversation is like talking to a brick wall.
> They didn’t do it alone, that doesn’t minimise their contribution, it contextualises them.
Oh, you very much minimised their contribution. From "when Proton came" (again, Proton came before Stadia) to "Stadia made gaming feasible on Linux" (when Proton made it feasible before Stadia)
> Also: Stadia ports of games were native, they did not use proton- it was architecture changes of the games themselves that made proton work better- not Google making proton itself function better.
So, Stadie games were Linux ports. But as a result of this there are still literally no Linux ports. None of the tech behind Stadia ever made it back into software behind Proton. And "native stadia ports" are somehow responsible for more games that target Windows and DirectX to run better via Proton
> That proton was running some games is a weird revisionist take
Funny to hear this coming from a revisionist. I literally provided you with links you carefully ignored
--- start quote ---
A look over the ProtonDB reports for June 2019, over 5.5K games reported to work with Steam Play
> You can even see this if you track the “gold” released games over years, some of the worst supported games for Proton are from 2015-16; before stadia but after game complexity started rocketing up with next game engines of the day.
Or because the actual heavy lifting that Valve did with Proton paid off, and not the nebulous "native ports" and code that never saw the light of day.
> because honestly this conversation is like talking to a brick wall.
Speaking from experience, Helldivers 2 and Monster Hunter Wilds both ran better on Linux from day one before any special fixes and still do - I'm not sure what "original design and battled testing" is worth or good for if the underlying Kernel and/or OS is a mess.
> and the scheduler on Windows is so poor compared to Linux ones that it’s even common for games to perform better on Linux than on Windows..
I play most of my games in a window and switch away a lot. A million years ago when I was still playing world of warcraft, the system overall was much more responsive on the same hardware with wow on wine on linux than with wow natively running on windows :)
> it’s actually Stadia that made linux gaming the most feasible
Stadia was the most predatory gaming offering aside from IAP games, sorry. Buy your games again on top of the subscription? Lose them when Google cancels the service? No thanks.
Nvidia's GeForce Now was a lot more honest. Pay for the GPU and streaming, access your owned games from Steam. I'm not using it any more so I don't know how honest they still are, but I did for like a year and it was fine(tm).
The fact that Stadia advanced wine compatibility is great, but technical reasons aren't the only reasons that make a service useful to your customers.
OP is talking about Google (Stadia) throwing money at the problem and incentivizing game engine companies to better support Linux. They’re not talking about pro or anticonsumer the tech was.
Stadia's impact on gaming in general is next to zero. And given that the vast majority of gaming on Linux is happening via Proton, its impact on gaming on Linux is similarly next to zero.
What games have you made to justify this statement?
I worked closely with productions using proprietary game engines, I feel qualified in stating that Stadia had an outsized impact on our development process in a way that helped proton succeed.
That you don’t see it as an end user, is exactly my point.
You don't have to be a chef to judge what's coming out of the kitchen.
What is the objective impact of Stadia which at its height had a whopping 307 titles [1]? At the time of writing ProtonDB lists 6806 titles as "platinum, works perfectly out of the box" and 4839 games as "gold, works perfectly after tweaks". Steam Deck alone has almost 5x the number of games with "verified" status [2].
What games are being made for Linux thanks to Stadia, and don't just target DirectX and run through Proton? How many Stadia games were ported to Linux thanks to Stadia?
Also, to put things into perspective. Proton was launched in 2018. Stadia was launched in 2019.
In 2019 there were already over 5000 games that worked on Proton. [3]
In 2022 there already were more games with verified status for Steam Deck than there were games for Stadia, and 8 times more games verified to work by users [4]. Stadia shutdown was announced half a year after the article at [4].
Stadia had zero impact on gaming in general and on gaming on Linux in particular as judged by the results and objective reality. Even the games you showed as examples don't support Linux, only target Windows, and are only playable on Linux through Proton [5]
> I feel qualified in stating that Stadia had an outsized impact on our development process in a way that helped proton succeed.
> That you don’t see it as an end user, is exactly my point.
It's strange to claim things like "when Proton came along" when Proton was there before Stadia and already had over 5k games working in the year when Stadia only just launched.
It's strange to claim outsized impact on development process when there are no outcomes targeting anything even remotely close to Linux development, with studios targeting Windows as they have always done.
It's strange to claim Stadia had outsized impact when none of the work translated into any games outside Stadia. When Stadia did not contribute any significant work to the tech that is running Proton. In 2022 they even started work on their own emulation layer that went nowhere and AFAIK never contributed to anything [6]
It's strange to claim that "it's actually Stadia that made Linux gaming feasible" when there's literally no visible or measurable impact anywhere for any claim you make. Beyond "just trust me".
You are literally arguing that your ignorance is as valid as my experience. And you’re arguing that you didn't see the impact; which was kinda my entire point - there was impact beyond what was visible that propelled Proton forward.
You don’t know how the sausage is made just because you ate a hotdog.
Maybe you should consider things more carefully before making yourself look like an idiot on the internet and simultaneously raising my blood pressure.
It's not really a failure.
Linux distribution and diverse ecosystem brings a level of complexity.
The only way to support it long term is to either having your team continuously update and release builds of the game to cater for that which is an impossible task to ask for a lot of studios.
The initial approach of runtimes did help but it's still has its limitation.
If now a studio just need to test their game under a runtime+proton the same way they would test a version of Windows to ensure it's working under Linux it's a win/win situation. Proton becomes the abstraction of the complex and diverse ecosystem of Linux which is both its strength and weakness.
Another solution would have been everybody using the exact same distribution which would have been way worse in my opinion.
And who knows, maybe one day Proton/Wine would be the Windows userland reference and Windows would just be an implementation of it :D
That’s not the goal though. The goal is to play games on Linux. If Valve’s goal was to end up with a Linux-specific graphics api for most games that run on Linux then they provably would have tried to do so.
When I was gaming on Linux, every game with a native version worked better using the Windows version in proton. I think the only exception was Factorio.
Gaming/WSL kept me on Windows for a lot of the last decade, however after Windows 10 became EOL'd and Windows started turning into ad/spyware I finally gave it up over a year ago after 25+ years on
Windows Desktops.
Anyway Linux is liberating, Fedora Desktop is great, no ads in the OS, a Software Store/Installer I actually like to use, curated by usefulness instead of scam Apps. All my Windows Steam Games I frequently use just worked, I have to login to X11 for 1 title (MK11), but everything else runs in the default Wayland desktop. Although I'll still check protondb.com before purchasing new games to make sure there'll be no issues. Thanks to Docker, JetBrains IDEs and most Daily Apps I use are cross-platform Desktop Web Apps (e.g. VS Code, Discord, Obsidian, etc) I was able to run everything I wanted to.
The command-line is also super charged in Linux starting with a GPU-accelerated Gnome terminal/ptyxis and Ghostty running Oh My Zsh that's enhanced with productivity tools like fzf, eza, bat, zoxide and starship. There's also awesome tools like lazydocker, lazygit, btop and neovim pushing the limits of what's possible in a terminal UI and distrobox which lets me easily run Ubuntu VMs to install experimental software without impacting my Fedora Desktop.
Image editors is the one area still lacking in Linux. On Windows I used Affinity Designer/Photo and Paint.NET for quick edits. On macOS I use Affinity & Pixelmator. On Linux we have to chose between Pinta (Paint.NET port), Krita and GIMP which are weaker and less intuitive alternatives. But with the new major release of GIMP 3 and having just discovered photopea.com things are starting to look up.
I hardly find anything interesting about command-line, I grew up in a time where the command line was the only way to interact with home computers, it fails on me the interest on staying stuck in up to early 1980's computing model.
Xerox PARC is the future many of us want to be in, not PDP-11 clones.
Weird flex, most commands, utilities, server software and remote tools are going to be run are going to be from the command-line. All our System Administration of remote servers uses the command-line as well since exclusively deploying to Linux for 10+ years.
Sure you can happily avoid the command-line with a Linux Desktop and GUI Apps, although as a developer I don't see how I could avoid using the terminal. Even on Windows I was using WSL a lot, it's just uncanny valley and slow compared to a real Linux terminal.
There are huge interoperability advantages to CLI and TUI tools. Composing them, using script(1) on them, etc, are much simpler than the same for GUI tools. They are also much easier to rapidly iterate on.
GUIs are very useful but they are not clearly better (or worse) than CLIs.
> Weird flex, most commands, utilities, server software and remote tools are going to be run are going to be from the command-line.
It's not a weird flex. Weird flex is this: "The command-line is also super charged in Linux starting with a GPU-accelerated Gnome terminal/ptyxis and Ghostty running Oh My Zsh" and then listing a bunch of obscure personal preference tools that follow trends du jour.
That’s not a flex, it requires no skill to install software, they’re just some of the better tools you can install to boost productivity in Linux terminals. I doubt they’re obscure to any Linux CLI user who spent time on improving the default OOB UX of bash terminals.
And you just alias them, so you can keep using the core utility names to use them.
Gaming on windows is fine, but there's no reason to use windows for anything else. Dual boot to linux for a better desktop and none of the crud that Windows 11 has in it.
I haven't been gaming since when there was huge gap between graphical possibilities and actual design (that is beginning of 3d era) - so I do not miss that. However I can see the decline in macOS, like pushing for 'apple intelligence', more and more restricting gatekeeper, iOS-ification of desktop (ie.: mentioned system settings), constant connections to AWS, etc.
But since I'm not gaming I cannot imagine going back to Windows. On the other hand I'm quite enjoying Linux...
> So why pay more for a lesser experience
...however, with few exceptions, I haven't used mouse in decade... and I haven't found anything like MBP's touchpad yet. Maybe I just need to do better research.
Not quite what you're after but if you want a fanless option that runs full linux and doesn't use much battery, the new argon 40 CM5 laptop that's being built looks like it could be viable as long as you'd be happy with that much of a drop in performance and a few pi based niggles (No USB C video, only one pcie lane for the SSD, etc.)
Yeah I stopped using spotlight a few years ago. I didn't really notice that I stopped using it until recently. It just became useless. I reorganised my stuff carefully so I know where I put it. I think that turned out to be more powerful than hoping a search engine over the top would be able to sift through the nuances.
... I'm not sure I've ever found any GUI system search reliable enough that I've used it on-purpose (though accidentally, sometimes) on Windows, Linux, or Mac. I always just use "find" and "grep" (and ripgrep when I remember that exists and realize I just grepped a lot and will be waiting for like an hour if I don't re-run the command with rg instead). Or nothing on Windows, which is fine because I haven't used Windows for anything but launching video games in about two decades.
Spotlight was bad back in the day, so I installed Alfred and started using that. Then Spotlight suddenly improved a lot, enough that it was usable for me, and I deleted Alfred. Then about five years ago something happened internally at Apple to the Spotlight team and it just got worse and worse and more difficult to use, making me regret deleting Alfred.
I wish Apple would just fix Spotlight. They don't seem to think it's worth fixing.
That is a good question. I like my dock uncluttered. I have it placed vertically on the left side, with only the apps I use every single day: Alacritty, Brave, Cursor, and Zoom. With Finder and Launchpad included, that's only six docked apps. Everything else I use Spotlight to open, so I feel the pain when the usability gets degraded or buggy.
Windows 11 LTSC one is quite good because it's so damn stupid. You can indeed hit start then just type what you want. Only does files though, by name, which is fine.
Windows 11 is beyond pale. It’s infuriatingly bad. But it could be a benefit if you do a bit of manual organizing and ignore most of its dumb features. Only use it for work, I will never use it at home.
I've only started using gui search since using fedora. the tracker search in the activities view is fast and finds files in the home folder by name pretty well. The only shame is that the pdf-content search doesn't work in the main search interface but only when searching in the file manager.
Apple (at least current leadership) is programmatically degrading it's products so people would buy a new one. Who expects anything good from such a team.
Regarding Spotlight, one thing that started happening for me on Sequioa was that Finder and other apps started getting very slow to react to file changes. For example, I can save a new file to a directory, and the Finder window takes maybe 10-20 seconds before the file shows up in the list. If I navigate to a different folder and then back, the file is there. I notice the same delay in apps like IntelliJ.
I could be wrong, but apparently Spotlight is the service that drives this kind of file system watching. I think macOS has a lower-level inotify-style file system event API, which should be unaffected, but Finder and these other apps apparently use Spotlight. I really wish I had a fix, because it's just crazy having to constantly "refresh" things.
This KILLS me. It's so frustrating. APFS is supposed to be great at deduping files and such, but in practice it seems like it really sucks. It's bad at both saving a file to the desktop and dumping a million npm files into a directory.
My favourite feature is when spotlight tells me that indexing is paused when I am searching for something.
You went through the effort to show some UI when something I am looking for may not be there because indexing is paused... but you didn't think to just unpause the indexing so that I can find it? I feel like I am being spit on, "Yeah, you not finding what you are looking for? I know, I'm not even trying"
I highly recommend using Alfred. I’ve been using it since before Spotlight came out, tried and then disabled Spotlight, and went back to Alfred. It’s extremely configurable but highly usable out of the box. Sort of like creating your own CLI shortcuts to open files, apps, copy things to the clipboard, etc.
I still use Quicksilver[1], the open source app that long predates Alfred and was the inspiration for it. I tried Alfred a few years ago but didn't see anything compelling enough to switch. Am I missing anything?
Same here. Spotlight used to be my everything, i.e. I never use the dock I would always use spotlight to launch applications or navigate to folders. Now it is littered with internet garbage, takes seconds to even return any results, and the results are always useless.
Who the hell thought integrating internet search is a good idea - because "aösldkfjalsdkfjalsdkfj" just as everything else is a valid search result in Spotlight now showing me "Search for aölsdkfjöalsdfjasdlfkj in Firefox"...
I worked at Apple Retail during the Snow Leopard launch. I think I still have a boxed disk somewhere, too. I remember it was not a product I had to sell to customers. People came in asking for it.
Another highlight of that job was selling a green iPod Nano to "John Locke" from LOST
Interesting that you think that of 7.5.3 — it worked, sure, but it could be painfully slow. System 6 was preferable as an OS — MultiFinder was better than 7, at least in the first couple iterations — but much of the software I needed demanded 7. 7.6.x was the first bright spot since 7.1 fixed much of what went wrong in 7.0, & there was a ton of waiting after that. 9 just chugged along for me, for the most part, which was nice.
Loved Snow Leopard too, & was shocked by how bad Lion was in comparison. Glad they got back on track after that.
You were right I forgot about 7.6.1. I think I had a an WiredInc mpeg video card server based on System 7.5.3 for a project. So it had a particular memory burn. I suppose I need up using System 9 since all life forms were supported by carbon.
I can't believe I'm saying it, but I agree with you about GNOME being my forever desktop. I used to really make fun of GNOME during the 2->3 transition, which seemed so profoundly misguided, but now I love it. I don't know if they've massively improved it or if my perspective has just changed with time.
It's not just you, the early GNOME 3 releases sucked. It has seen a lot of gradual improvement over time. Of course there are reasonable alternatives also, such as Xfce, MATE or Cinnamon. (And these three 'alternative' desktops have also edged closer over time, sharing more and more of the underlying tech stack even as GNOME itself has continued to develop in a rather seperate direction).
Unified button that disguises as two different icons hiding other useful options
You can only cycle windows in one direction even if you try to do some remapping
Choosing keyboard languages hides a lot of options. Once you understand you need to click on English US to see more detailed options then you get them all, UK, Canadian...
Then it's unclear which keyboard layout is currently selected and how to select one from the list you made.
I can't fathom how a DE whose is all about human machine interface guidelines whatever and supposed to be the epitome of UX can't figure out basic stuff about discoverability and clarity
True ! So why can I only remap cycling window in one direction and not the other ... ?
The volume and power icons on the top right is actually one button and hides other option like screen lightning volume and wifi etc.
If at list they had made a three vertical dots/stacked bars and is the convention for hamburger menus...
From what I heard GNOME devs do not like change and it sucks to be a GNOME extension developer, a quick google search seems to confirm that so it casts some doubt about them up-streaming any of them but maybe you know better. Has it ever happened to other extensions ?
Haven't really used MacOS or iOS more that five minutes so I can only trust you on that.
On the other hand for example, it is very easy to remap CapsLock to Escape on MacOs. Just go to Setting --> Keyboard and you easily find the option. GNOME ? No, not in settings. Wait I have to use an app called gnome-tweak ? Ok it's in "Advanced keyboard otions" --> Opens a big list of loosely classified options. Oh well it was in miscellaneous category.
I can believe that its easy to bounce off software because of a million paper cuts. But the problem with them trying to address every one of those proactively is that GNOME is a huge undertaking and they do their best to move at a fairly slow pace (now, after the 3 transition, which was akin to ripping a bandaid off ... go fast, piss the person off, but then the bandaid is gone).
I don't know if the CapsLock -> Escape switch is on a roadmap somewhere, but that is a little bananas. That said, my partner comfortably uses GNOME every day to browse the web and manage some files. Has she EVER wondered how to remap CapsLock? No. The people who do want to? Google can give you the answer pretty quickly. Not saying it's good UX, but GNOME balances a lot of use cases, and as this thread suggests, I think they've actually (with a LOT of complaining from engineers and power users) kept that balance pretty damn well to the point where I haven't been surprised by GNOME is a long time, and seems to slowly and progressively get better.
And yes, whoever jumps in here with their own papercut story, I know there is pain in not being the primary audience for software. But honestly, at least I'm in the same Venn diagram with my partner. The primary audience for macOS or iOS now appears to be ... I don't even know anymore. Used to be content creators, now it seems like even Apple doesn't actually know who uses their computers.
Wow, I feel like I almost could have written this except I prefer Plasma/KDE to GNOME. I use Linux + Mac laptops somewhat interchangeably since 2012, and have also seen the marked decline in quality. In fact, it seems like Linux has gotten better at almost the same pace (or maybe a bit faster) than macOS has gotten worse.
The things that most frustrate me about Macs is that they've violated the never spoken but always expected "it just works" in so many ways. Things like how Thunderbolt Displays containing a USB hub which are Apple-certified handle re-connection to a Macbook, should "just work", but require fiddling every time. That's just one of numerous examples I could come up with.
Apple historically was probably the best company in the world in understanding the full depth of what "User Experience" means, and it seems like they've really retreated from this position and are regressing to the mean.
At one point a few years ago, Spotlight improved enough that I could use it instead of relying on Alfred. So I deleted Alfred, and whaddaya know...a few years later Spotlight got worse and worse, making me regret that move.
My biggest annoyance with recent macOS versions that most QuickLook plugins stopped working. Apparently one could re-develop them with their new framework-of-the-day, but I have no doubt a lion's share of what I'm using will just become abandonware.
Just noticed that I was sharing my entire Safari, Spotlight and Siri search history in that menu. Why is that setting in Spotlight settings and not under Privacy/Analytics?
I've found that kind of thing is often caused by a damaged preferences file. The easy way to check that is to make another user account, and see if it happens there too.
Spotlight is unbelievable bad, especially on iOS. If I type a substring of the name of an installed app, it should find it effectively instantly (say, within 1-2 frames of the input showing up). Instead, it finds it sometimes. On occasion I need to hit backspace (removing a letter that should match) to get it to find it.
I struggle to imagine the software design that works so poorly.
I've yet to find a decent implementation of search-as-you-type anywhere, not just Spotlight. I have that same issue on Firefox, and with Windows Search, for example.
And it makes no sense whatsoever. If "foo" matches "foobar", so should "foob". I honestly don't know how the hell can they still f up such a simple piece of technology in 2025.
Windows 7 start menu search was always reliable and had predictable behavior from my experience. It can be done, just that modern software engineers' skills and career incentives no longer permit it.
Spotlight was never useful, because of an absurd and glaring design defect: It doesn't show you WHERE it found stuff. There's no path shown with hits. Same blunder in Finder's search, and you can't even optionally add "path" as a column. WTF.
So... when the hits include six identically-named files, you can't eliminate ones that you know are wrong (on a backup volume or whatever). The level of stupidity here is just mind-boggling.
In all fairness, you do need to hold down the command key to show the file location in Sequoia. It is an interesting default behavior to pretend the files location doesn't exist, mobile-centric.
No you don’t. In Finder search results, the path is always shown at the bottom. For regular Finder windows, you can optionally show the path with “View -> Show Path Bar”
In all fairness, secret hotkey BS may as well not exist. Are you supposed to mash every modifier key and every combination thereof on every screen and in every menu, looking for hidden goodies?
we have simplified the interface to just one home button and the screen interface, as well as the volume up/volume down key.
To select, just press on the item.
To hover, press and hold for at least 2 seconds.
To get a list of options, press and hold for at least 2.5 seconds, but not more than 3.5 seconds.
To delete, press and hold for 3.6 seconds, but not longer than 3.9 seconds.
To save, press and hold for 4.1 seconds. Pressing and holding for exactly 4.0 seconds activates the archive action. Pressing and holding for 4.2 or more seconds sends the item to the blocked list.
To retrieve the list of items in the blocked list, press and hold and simultaneously press the volume up and volume down key.
To delete all items in the block list, press and hold and simultaneously press the volume up key only.
To completely reset your device, press and hold and simultaneously press the volume down key only, whilst holding the device in a completely vertical plane, and rotating clock-wise and counter-clockwise, smoothly, at precise 2.35619 radians every 30 seconds.
To trigger the emergency call feature, drop the device at an acceleration of no less than 9.6m/s and no more than 9.7m/s
No: you are supposed to read the documentation to learn about power user features. Microsoft also doesn’t shove the advanced keyboard shortcuts in your face; you need to read the manual to learn stuff like this.
Is it, though? Most people don’t really have a notion of the file system, or hierarchical file structures. They drop files onto their desktop, or keep them in the downloads folder. Just ask a parent or your next-door neighbour.
That’s a bit of a problem when discussing problems of normal users with power users, because they don’t even realise how what they’re doing is actually not what normies do.
I’m inclined to agree that hotkeys in MacOS are hard to discover, but cluttering the interface with stuff many users simply do not need cannot be the correct answer.
Where? And how is that option displayed to the user?
I also just tried it in Spotlight and Finder, and it did nothing. Which I consider a relief, because undiscoverable bullshit is worse than the feature not existing.
macOS and iPadOS are full of those undiscoverable "if you do this combination of buttons/swipes while at full moon, something happens". As a Mac user not by choice (work issued) I hate how impossible to discover these are.
As a Mac/iOS/iPadOS user it seems that it’s almost mandatory to watch each Keynote / product announcement video if you want to keep up with new features. Lots of cool features that I only knew about by watching those videos that are completely undiscoverable otherwise.
These kinds of shortcuts are part of Apple software as a whole, and apparently have been a thing since at least OSX. These behaviors were supposed to be covered in the documentation, but I don't know how true this is nowadays.
Special mention to all text input fields in macOS having Emacs-style shortcuts.
It goes back further than that. I remember being able to buy key-combo cheat cards for System 7, and I have no reason to think the shortcuts they covered wouldn't also have been present in System 6.
I agree that discoverability could be better, but macOS has pretty consistently had hidden power user shortcuts and modifiers, to keep the basic workflow streamlined/simple for those who don't need it.
Seeing where stuff is in a search function is not a "power user" feature; it's the whole point of what you're doing.
And I don't buy the "keeping things simple" excuse for secret hotkeys in other areas. Falling back on gimmicks like undisplayed hotkeys and "long presses" and "gestures" is lazy abandonment of the design task.
I hate this "saving the user from complexity" lie. It's hypocritical: The "non-power" user isn't going to go looking for these options in the first place.
Finder search is a great example. A "non-power" user isn't going to right-click on the column headings in the results and try to add "path" as a column. So how does it help that user to deny everyone else the ability to add it?
Apple mocked IBM for needing a thick user manual back in the day. To suggest that anyone (especially anyone on this site) should have to read documentation to use perform a basic file search (in a GUI, no less) is apologism to the extreme.
And now spotlight defaults to the whole computer even when I start a search within a folder... for items in the folder... Turned to garbage sometime in the last ~18-24 months.
I personally think that it is reasonable to "want" an Apple notebook. They have great hardware, great battery life and an ecosystem where every device integrates. Only on macOS you can nicely develop software for iOS. Furthermore most vendors release software for macOS, while they don't for Linux (not only Adobe). BTW the apps I miss most on Linux is the Preview App and Apple Mail.
However I'm done with Apple. I think it's a decision - not "reasoning". That decision takes time and is painful. It's also a decision specifically against "the best" ecosystem available in favor of something "ok".
Not only they repeatedly disappointed my expectations - they just suck as a company (in my opinion). It's not about being less innovative for decreasing software quality, they have done so much for the market, that I think GNOME wouldn't even exist as it is without them... Its about sealing off every inch of their software and hardware they can. No repair without paying... Making RAM and SSD upgrades ridiculously expensive, you cannot even put default NVMe drives into a mac mini - everything is proprietary. Even their sensors have serial numbers to prevent hibernating if you change them out without "hacking" the firmware.
Hardware-wise I have high hopes for framework working with AMD - although they did not address the issues I'd suggest (speakers, lpcamm2), they're constantly improving without breaking their promises. This is hopefully not going to change when they get bigger.
OS-wise I'll stay on Linux. After a long journey going from Ubuntu to Debian to Fedora using GNOME, KDE and even NixOS with Hyprland for a short period, I gained enough knowledge required to really enjoy Linux. System76 is working on COSMIC, which could be pretty amazing, once it is released.
In case anyone would like to try my current Linux config, I'm constantly working on an "install everything" script (pretty early stage):
Yeah... probably. I forgot to mention that Apple computers are a pretty good deal if you are looking for an AI / LLM experimentation machine due to unified RAM which nearly translates 1:1 into VRAM.
Indeed. I complained that Apple design gets a free pass while being haunted by Steve from beyond the grave for a decade. Your comments resemble my habits except rusted sway right into cosmic desktop alpha and done.
What is wrong with an AMD Ryzen 9 with 16 physical cores? If you need more and you have a virtually unlimited budget, then Ryzen Threadripper is even better. Also: Is Asahi Linux an option for you?
Did you know, you can set your wallpapers to be continuously updating and make macs use terabytes of your network in hours or days depending on speed?
https://discussions.apple.com/thread/255329956
Apple's software quality (either in terms of polish or just plain QA) has steadily decreased
I think the decline of software went hand-in-hand with the decline of the native indie Mac app. They still exist, but when I started with the Mac (2007), there was a very rich ecosystem of native Mac apps. Most stood head and shoulders above their Linux and Windows counterparts.
Apple has nearly destroyed that ecosystem with: race-to-the-bottom pricing incited by the App Store; general neglect of the Mac platform (especially between ~2016 and Apple Silicon); and a messy reactionary toolkit story with Catalyst, SwiftUI, etc. The new toolkits seem to imply that Apple says that it's the end of AppKit, but most SwiftUI applications are noticeably worse.
With their messy toolkit story and general neglect, developers have started using Electron more and more. Sure, part of the popularity is cost savings, since Electron apps can be used on multiple platforms. But part of it is also that a Catalyst or SwiftUI app is not going to provide much more over an Electron app. They will also feel weirdly out of place and you become dependent on Apple working out quirks in SwiftUI. E.g. 1Password tried SwiftUI for their Mac app, but decided in the end that it was an uphill battle and switched to Electron on Mac instead.
I recently bought a ThinkPad to use besides my MacBook. Switching is much easier than 10 or 15 years ago, since 80% of the apps that I use most frequently (Slack, Obsidian, 1Password, etc.) are Electron anyway. Even fingerprint unlocking works in 1Password. I was vehemently anti-electron and still don't like it a lot, but I am happy that it makes moving to a non-Apple platform much easier.
I think most of this is just downstream of the Mac being eclipsed by the iPhone in terms of Apple’s revenue. The Mac just isn’t critical to Apple’s business like it was in 2009 when Snow Leopard came out. They would have started development on SL in 2008, when the iPhone was still a fairly niche product and there wasn’t even an App Store.
Now, ios gets the executive attention and it will generally get the best developers assigned to it, and the Mac has to live with the scraps.
Yeah I think this is the one, in terms of number of users, revenue, etc. The iPhone is more than 50% of their revenue, Mac is only ~8. Lower volume and higher price, but it doesn't come anywhere near their phone. Same with tablets, although they share an app revenue income stream with the iphone which makes up for the difference in hardware sales.
1Password had a beautiful native Mac app that works to this day. Even assuming SwiftUI is actually bad, why did they have to migrate at all? What was wrong with the existing app?
I'm not disagreeing with the opinions on Apple software quality, but I think the 1Password case is more down to their taking of VC money and having to give (JS) devs some busywork to rebuild something that worked perfectly well.
1Password is also now subscription only and online only. Gone are the days of a forever license and fully offline encrypted database allowing for 3rd party syncing via iCloud or others. The death of their old app went hand in hand with their race to the bottom subscription payment VC backed ecosystem. It's only time until they suffer a breach like everyone else.
> I recently bought a ThinkPad to use besides my MacBook.
I'm on the same boat here. Something is driving me away from my MacBook M1(Pro? Don't even know). I have a gut feeling that it's macOS but can't really put a finger on it yet.
Bought a heavily used ThinkPad T480s (from 2018) and replaced almost every replaceable part of it, including the screen. Being able to replace many parts easily is a nice touch since I am using MacBooks since 2007 exclusively. Guess that's why I somehow overdid it here.
Slammed Pop!_OS 22.04 on it and I'm very pleased with the result. The first Linux desktop I actually enjoy since trying SuSE 5-something.
Pain points are teams (running in browser), bad audio quality with AirPods when using the microphone and cpu speed and heat. I guess one has to stop using Apple silicon in laptops to realize how amazing these processors are.
Intel CPUs from that era were quite bad and everyone has upped their ante since then. I was thinking about getting a second hand from ~2021-2022, but my wife convinced me to get a new one, so I got a Gen 5 T14 AMD. It has a Ryzen 7 Pro 8840U and I rarely hear the fans, mostly only when Nix has to rebuild some large packages (running NixOS unstable-small).
> As someone who's been a Mac user since System 6 and has been consistently using Macs alongside PCs _daily_ for over 20 years I can say that Apple's software quality (either in terms of polish or just plain QA) has steadily decreased.
> As someone who's been a Mac user since System 6 and has been consistently using Macs alongside PCs _daily_ for over 20 years I can say that Apple's software quality (either in terms of polish or just plain QA) has steadily decreased.
Similar for me but started in system 7.
It’s lucky for Apple that Windows has got worse faster.
I do love Gnome. If only we had hardware to run it :/.
I'm stuck on a MBP because it's the only laptop with a great screen, speakers, and battery life. Meanwhile my keyboard keys keep getting stuck after a year of usage, and OSX is garbage. Soon as there is similar hardware I can load Linux on, I'll be insta-switching.
AMD AI Max 395 (superb name) proved that x86 can get to Apple silicon perf. (even not that far in power efficiency), but there seems to be 0 devices from non-trash brands (I am not buying ASUS or HP).
I would love to finally get out of Apple ecosystem, I just don't have any decent alternatives right now. Hopefully next year.
I'm going to purchase a framework just because I value repairability. And honestly, before the m1 macbook I was using a t480s, and I'm okay with compromising on hardware, esp. having been burned with the 2016 butterfly macbook. Apart from the haptic touchpad I wouldn't miss much, other makers are finally ditching low resolution 16:9 screens and you can even find nice oleds. I'm mostly missing the polished software that's only available on macos (things like carbon copy cloner or pixelmator). But with my m1 having degraded battery and having to send it off for a week or two to the nearest service center just to get a new battery, the prospect of a repairable laptop like framework where I can just order a new battery and replace it myself is looking all the more enticing.
Another example was High Sierra. They completely swapped out the file system on that release, focusing primarily on under-the-hood changes, and imo was also one of the most stable macOS releases to date.
This is an interesting idea, and I am actually curious what Apple is going to do going forward. A "Snow Leopard"-esque release would be nice, but I think what would be better is an LTS release. Historically, you get a new Mac and you usually only get 5-6 years before they drop your model from the latest release. This has always made some sense to me, as after 4-6 years, you do start to feel it.
I bought an M1 Max that is now almost 4 years old and it still feels new to me. I can't really imagine a change that would happen in the next 2 years that would make this thing feel slow where an M3 would feel sufficient, so I'm curious to see if Apple really does just go hardcore on forced obsolescence going forward. I have a few M series devies now, from M1 to M3, and I honestly cannot tell the difference other than export times for video.
I can imagine some kind of architecture change that might come with an M6 or something that would force an upgrade path, but I can't see any reason other than just forcing upgrades to drop support between M1-M5. Maybe if there is a really hard push next year into 8K video? Never even tried to edit 8K, so I don't know. I'm guessing an M1 might feel sluggish?
I don't feel like they ever used forced obsolescence with Mac's. When they dropped support for the latest OS on your machine it was usually because it couldn't run it. I recently updated some older Mac's and even a couple of OS's before support was dropped things got really sluggish. I imagine with the Apple Silicon machines the OS support will stretch longer than it has on the Intel ones. Maybe the higher prices are a hint they expect people to keep the machines in use for longer than before.
They need to somehow start marketing effectively to gamers, because the GPU in your M1 Max is shit. Sure, it’s fine for mostly-2D UIs and the occasional WebGL widget, but for AAA gaming it’s just dogshit.
'Gaming laptops' with more powerful GPUs are generally awful, though. Even ignoring the state of Win11.
Yes, they can theoretically perform better, but only when plugged into mains power, and creating so much heat and fan noise that the experience really isn't good.
Don't think there's anything out there that will outperform the GPU of an M-series Mac without consuming way more power and producing problematic levels of heat+noise.
Sure, but this is another avenue to onboard people to the upgrade train. Sure your display is great, your CPU is great, the speakers are great. But the AAA graphics scale up every year and there are often big performance cliffs for new features on old hardware.
> I think what would be better is an LTS release. Historically, you get a new Mac and you usually only get 5-6 years before they drop your model from the latest release
In fairness, Apple to do tend to continue to release critical security patches for older versions.
I suspect that it will be AI features that push Apple into deprecating older hardware. But I also hope that the M series hardware will be supported a bit longer than the intel hardware was. Time will tell.
> I bought an M1 Max that is now almost 4 years old and it still feels new to me.
How are the keycaps doing? Mine looked awful after about 2 years of relatively light use, developing really obvious ugly shiny patches (particularly bad on the space bar), quite a letdown on an otherwise great machine.
(Realised that you can actually buy replacements and swap them yourself, via the self-service repair store, so have replaced them once, but am starting to notice shiny patches again on the new set)
Not OP but have the same Mac. Every key is shiny. Doesn't really bother me though because I touch type. Also clearly I favor hitting space with my right hand because only the right side is shiny.
If you have AppleCare they will basically rebuild your MacBook for ~$200. I got MBP M1 Max usb ports and top case replaced and a bunch of other stuff I didn’t even ask for but they replaced with new stuff. Felt like a new machine when I got it back.
Still better than the butterfly debacle of 2016-2019. I have one for work that spends 99.9% of its life docked to a real keyboard and it still has keys that only work sporadically. Some of these keys probably have < 10,000 actuations on them.
Trying to use Wan2.1 to generate AI video or other various LLM or StableDiffusion style stuff is slow compared to other other platforms. I don't know how much of that is because the code is not optimized for M1+ Max (Activity Monitor shows lots of GPU usage) or how much of it is it's just not up to the competition. Friends on 4070 Windows PC are getting results many X faster and 4070 perf iss not even close to 4090
I don't have any Macs or iPhones that can even run the latest software anymore. My absolute newest Mac is stuck on Ventura 13.7. On the other hand, I can get the bleeding edge version of any Linux distribution out there and run it on decades-old hardware.
Unfortunately, “decades old hardware” doesn’t give me the combination of speed, quietness, battery life and the ability to use my laptop on my lap without so much heat that it puts me at risk for never having any little Scarfaces.
Using an x86 laptop in 2025 is like using a flip phone.
That way I can go and buy a MacBook Pro (13"?), or a MacBook Max Pro (15"), or a MacBook Mini (11"), or a normal iPad Mini Ultra, or an iMac Mini (21"?), or a Watch Pro, or a Mac Max Ultra etc.
I agree. It seems ridiculous that an app like Messages is considered so much part of the OS given what it does. I don’t use it, I don’t care about it, but it seems like it could be a regular app that updates independently of the OS, along with Maps, Notes, and so on. So many macOS “upgrades” nowadays seem to be Apple tinkering with such apps rather than the actual OS experience.
Why do you want to destroy computing for everybody else just to make a very small group of hackers pleased? Can't you be satisfied with Linux or Windows or BSD? Let normal people have at least one platform that is usable for them.
How is letting users to disable bloat a bad feature?
For example, can you remove Chess from MacOS? Nope! Why? What I found on Reddit, it seems because it's integral part of MacOS somehow and I am a bad person for even asking, somehow.
I did said similar thing weeks ago regarding all shenanigans with W11: every software should include two paths of installation/OOBE: default "express" where vendor shoves you "the experience" and customized "expert" where you select features YOU want. And either way allows you to change system afterwards. We had that in Windows years ago but then it was removed; some Linux distributions do offer package selection beyond the default set.
There's no need for a separated core version - just give back control to the user. But honestly, I don't know what would need to happen so we could get it - it feels like it's a lost cause against corporations. There's of course Apple-EU situation where you can remove applications, set defaults, install additional app stores but this is still limited to that market and happen way too late and too slow.
> I am not suggesting Apple has fallen behind Windows or Android. Changing a setting on Windows 11 can often involve a journey through three or four different interface designs, artifacts of half-implemented changes dating back to the last century. Whenever I find myself stuck outside of Appleland, I am eager to return “home,” flaws and all.
Hard agree with this. I sometimes have to boot up a windows laptop to play Minecraft with the kiddo, and it never stops reminding me how little I know about Windows now, how counter-intuitive everything is, how everything feels designed for a user whose mind I cannot comprehend.
Also, every time you run something in Windows (whether it's part of the OS or an App) it can be a trip down memory lane, UI-wise. Oooh, this dialog is 2015 vintage! This dialog is styled like Windows 8! This one is from the XP era! Ohh, and that rarified dialog has controls that have not been changed since Windows 95!
If you want a super bad audio-related journey, try fixing external speakers connected to a Linux box. It's abysmal, and 99% of it can only be done via the CLI. Nothing wrong with that... but for something so normal I expected more ease-of-use.
Hey, if we're already complaining about Microsoft products, can someone explain why the Bedrock and Java versions of Minecraft have not been made cross-compatible in the TEN YEARS since the Mojang acquisition?
(... speaking as another dad just trying to play with my kid.)
What does cross compatible mean in this context? They are two different games written in two different languages. I mean, they look like they are the same game, but they are not. Making one compatible with the other is a Herculean task. If not impossible.
I’d imagine mostly due to a lack of incentive on microsoft’s part. Like minecraft is literally the biggest video game to ever exist with, making 2 entirely separate code bases work while keeping all the features the same and preserving compatibility with over a decades worth of mods just so the mostly separate java and bedrock communities can play with each other is just not worth the risk. So many people play minecraft in so many different ways means that making even minor changes in gameplay can be huge sources of controversy, let alone major infrastructure changes.
They still exist separately today because the modding scene is completely different for them. Minecraft Java is the original and has a huge modding community based on decompiling and patching the game. Those mods are all incompatible with Bedrock because Bedrock is a separate reimplementation of the game for performance or whatever.
I... think so? Whichever one works with Microsoft Realms, which is the $2/month solution I settled on after somewhat-getting a self hosted server to run for a little bit on my desktop.
I figured that I make a six-figure salary as a software developer, I can afford $2/month so that I don't have to fucking become a sysadmin for a game server my child depends on.
There are two editions, Java and Bedrock. Java is the original, available on PC and Mac, and supports programming-like technical play and mods. Bedrock is Microsoft’s reimplementation, available on all devices except Mac, and supports emotes and microtransactions. Other than that they’re largely the same game, and buying either gives you both versions. Realms supports both, but a server is one or the other, not both. There are also other managed hosting providers for Minecraft (both versions), but Realms is probably easier and cheaper for you. Java version has performance problems, but mostly because Microsoft’s code is inefficient, there are a few mods (also written in Java) that everybody uses to fix performance without affecting gameplay.
Search feels to me like a good compromise between memorizing terminal commands (including the correct set of parameters to do what you want) and navigating through a UI to find what you're looking for.
Search is fine as a one-off thing, but if you repeatedly have to use search to find some common setting, that's a clear UX fail.
To be fair, it's hard to say whether the Settings app is more broken in Windows or macOS these days. I think I'd have to give the crown to macOS here on account of search itself being more broken.
Why is it a UI fail? Honestly search as the default way of going to settings is my favorite development in modern OS design, I no longer need to memorize 3-6 deep menu trees to find a trivial setting.
For example:
I prefer keeping my hands on the keyboard, and typing cmd+space followed by mouse is so much faster than finding the right pixels to click through in menu trees when I want to adjust my mouse sensitivity.
I didn't say that search itself is a UX fail. It's not; it's great!
The UI fail is if search is required to find the setting every time you need it, because categorization and/or navigation is broken otherwise.
As to keeping your hands on the keyboard, that's an argument for having proper keyboard support in any UI, complete with hotkeys and shortcuts. The big difference between these and search is that the former is (if properly done), consistent and predictable. So e.g. when the app adds new things in the future, your existing key sequences will still do the same thing they did before.
To take your specific examples, if I do Cmd+Space, "mouse", Enter on my system, it will bring up LinearMouse, not system mouse settings.
Disagree with this. I use the search for everything. It’s just so much quicker than even a well designed UI.
On my iphone, I have one page of apps, everything else in the app drawer, and use the search all the time. It often gets what I want in one or two chars.
I admit I honestly have no idea where the system settings are located as I haven't pressed the start button in ages, but the same applies to MacOS as I would use spotlight there as well.
Using search as a UI means you can only find things that you know exist, but there are plenty of important settings that I've only discovered by actually navigating through the UI.
The point of this comment thread is that important Windows settings are scattered throughout many different interfaces beyond just the Settings app, and you can never be sure where to find what you're looking for, which results in a poor user experience. Off the top of my head, you have the Settings app, Control Panel, Device Manager, System Configuration, and Network and Sharing Center.
then Mac fails as hard as windows. there’s a reason search exists in the settings app on both MacOS and iOS. and there are plenty of settings that require “default write …” or editing some plist file or worse
> You can access most settings by Windows + "yourquery".
The search doesn't even work all the time. Sometimes it won't do fuzzy search, sometimes typing "bluetooth settings" will do a Bing search, some other time it will open a PDF, and so on.
I recently discovered that I can change audio settings on a mac by using the opt+volume shortcut and it takes me directly to the sound panel. Now if I could only make it stay on the built-in microphone instead of always switching to the worse sounding airpods one.
It's fine if you stay away from the consumer releases. Windows 11 LTSC (based on 24H2) feels like windows 7. Most of the stuff you had to futz with powertoys and GPOs back then. That hasn't changed. I quite like it. It has been utterly boring compared to my recent Apple experiences.
I disagree with that. As an occasional user of MacOS, the new Settings app is quite bewildering. There are just as many dials as in Windows and sometimes requires a trip to ChatGPT.
And for reasons I don't understand, why is the window itself not resizable?
I'm a Windows fan (I actually really like 11) so I'm a bit biased, but I just dove back into macOS since 2014 and the settings app is truly terrible. The built in search barely works and the layout is so damn confusing. God forbid I install some remote desktop software, now I have to go to accessibility settings 5 simes and approve some permission that is strategically buried for only what I can tell as a way to thwart "normies" from enabling something via obfuscation.
It would be fine if the settings available were actually useful or at least could bring me to some tool that does it better. I get no meaningful report of what's eating my batter and why every time I open my MacBook it's dead. And if I want to change the actual resolution of my display I'm given just a list of scaling options pretending to be resolutions. Oh, want to set a specific resolution or refresh rate? You have to do some stupid kinger king foo of option control something _before_ you click on this dialog. I get the criticism about the Windows settings app and legacy power tools (I think this has largely been solved anyway), at least they exist and allow me some iota of control over my computer
It is resizable vertically but not horizontally as it doesn't make sense to resize the window horizontally considering the content of settings details panel (the right part of the settings window), you would end up with a lot of empty space if you were able to resize it horizontally.
You could say the same thing about the Windows Settings app, but it resizes in every way and it's very much size adaptable. In other words, UI components resize or become visible/invisible depending on the width.
Except for when the placement of the icon strip with the trashcan symbol changes to the bottom of the context menu because of the location of the context menu on the screen. Bonkers. No idea why the UI committee would’ve okayed that one.
As someone who stopped using windows about 7 years ago, and only recently used it last weekend, my eyes probably glossed over the fact that some buttons were laid out
horizontally.
I was fully braced for Windows 11 being awful when I installed it recently but that hasn't been my experience at all. If anything it's just a slightly more polished Windows 10.
Probably helps that I installed the IoT LTSC version, but still, apart from the task bar being stupidly in the middle (thankfully there's an option to move it to the left), I've had zero issues.
I even added a network printer and it found it quickly, and added it quickly and successfully, which is a feat I don't think I've seen happen on any OS ever.
The context menu is a clear improvement on the old one (which you can still get to with one click).
Every article about some issue with Apple MUST also include an anecdote about how you couldn't use Windows one time and how it's still worse than Mac.
It's the rule lest someone think you made a bad decision and you're regretting it. Even though it's an OS targeted for your grandmother, you must not let them see weakness.
At this point it's a joke. Either critique Apple or admit you can't without also bringing up some other OS. It's weird.
TL;DR What people remember fondly is not Mac OS X 10.6.0, which was in fact very buggy, and buggier than 10.5.8, but rather later versions of Snow Leopard after almost 2 full years of bug fixes.
The yearly release cycle is the problem. Apple needs "another Snow Leopard" only in the sense that I mentioned above, "almost 2 full years of bug fixes", although at this point, Apple has more than 2 years of technical debt.
Thank you, the nostalgia for a 15-year-old OS release, which absolutely was not great out of gate, is strange.
My recommendation for people who don't absolutely need the latest features: Upgrade to the previous version of macOS when the new version is released. Sequoia is incredibly reliable 7 (soon to be 8) updates in.
I disagree. From the article: "The same year Apple launched the iPhone, it unveiled a massive upgrade to Mac OS X known as Leopard, sporting “300 New Features.” Two years later, it did something almost unheard of: it released Snow Leopard, an upgrade all about how little it added and how much it took away. Apple needs to make it snow again. Snow Leopard did what it was made to do. It was one of the most solid software releases Apple ever put out." This gives the impression that it was solid out of the gate, which it was not. And the next paragraph specifically mentions "2009’s Snow Leopard". But later Snow Leopard releases were in 2011.
Grandempire is right on my overall sense in the piece, though perhaps I should have made its ore explicit. I actually faired quite well with 10.6.0. But, the lack of push for a yearly set of headlining features did allow the OS to age quite well in the years after, too. It's the drumbeat of what 10 stunning new features will be unveiled each WWDC for each platform that means past features rarely get the continued polishing they need to shine.
I feel like they're trying to build too many platforms most of which have become quite large. macOS, iOS, iPad OS, visionOS, watchOS, tvOS. The fact all of these systems are quite tightly linked in terms of features/syncing makes it difficult to navigate. If you want to ship every single year you need more developers, but that might make the collaboration between the systems more difficult. They need to move away from the one year cycle. It's a stupidly short period of time to ship a whole OS (or 6 whole OS's). If you want to keep them all in sync switch to two year cycles and decouple some of the apps from the core OS (e.g. Music, Safari, etc) so they can be updated as necessary outside of the cycle.
The platform teams at Apple don't really work that way. My (limited) understanding is that they share a fair amount of core code, but each went their own way for a while and have recently started getting nudged back together from a UI perspective -- unfortunately, the iOS style guide seems to have won, and many decades of desktop UX is being thrown out with the bath water.
I'm done with macOS, I've migrated to Linux for my general purpose computing. With every new release of macOS, Gatekeeper is becoming harder and harder to bypass, increasing Apple's control over what software can be run on macOS, forcing apps to be signed with an Apple Developer ID. While I'm happy they are taking security seriously, I'm seriously creeped out that macOS sends hashes of every executable I run to their cloud. It's starting to feel like a broader move away from the openness of personal computing and towards a more controlled, appliance-like software experience.
When Sequoia eliminated the ability to override Gatekeeper by control-clicking, it became clear to me that Apple is now employing a frog boiling strategy towards their ultimate goal -- more control of the software you can run on their hardware.
My group makes a custom executable to reflash a hardware device we produce. We build it for Linux and Darwin.
Trying to get the program to work with our Mac users has become harder and harder. These are all internal developers.
Enabling developer mode and allowing Terminal execution isn't enough. Disabling the quarantine bit works - sometimes - but now we're getting automated nastygrams from corporate IT threatening to kick the laptops off the network. I'm exhausted. The emergency workaround, which I tell nobody about, is way less secure than if they just let us run our own software on our own computer.
I understand that you're doing it on principle, but for a software development team, 99$/year is a really minuscule price to pay to be able to build / notarise / distribute software.
Developers pay exorbitant amount of money for much lesser value, and the idea of putting your teammates at risk to stick it to apple is kind of sad bordering with negligence from a business POV.
Adding signing as a requirement can easily make what was once a very simple distribution mechanism into something much more complex - now you need manage signing certificates and keys to be able to build your thing.
You seem to be comparing a single dev sending apps to the world vs a corporate team pushing to employees (if I get parent's case right).
In most cases, just involving account management makes the corporate case 10x more of a PITA. Doing things in a corporate environment is a different game altogether.
In contrast to this point, as long as I use Xcode and do the same thing I've always done allowing it to manage provisioning and everything else, I don't have a problem. However, I want to use CI/CD. Have you seen what kind of access you have to give fastlane? It's pretty wild. And even after giving it the keys to the kingdom, it still didn't work. Integrating apple code signing with CI/CD is really hard, full of very strange error messages and incantations to make it "work".
I don't know about fastlane, since my CI/CD is just a shell script, and signing and notarising is as hard as (checking the script) running `codesign ...` followed by `notarytool submit ... --wait`
Yes, you need to put keys on the build server for the "Developer ID Application" (which is what you need to distribute apps outside of AppStore) signature to work.
You do not need to give any special access to anything else beyond that.
Anyway, it is indeed more difficult than cross-build for Darwin from linux and call it a day.
Code signing is absolutely disgusting practically and philosophically. It has very reasonable and good intent behind it, but the practical implementations cause great suffering and sadness both for developers (cert management, cost, tools) and end-users (freedom of computing).
Do you distribute OSS software which requires notarizing? If so, have you found a way to let the community build the software without a paid developer account? I would be very interested in a solution which allows OSS development, relying on protected APIs without requiring that anyone who builds the app to have a paid developer account.
The principle is what matters. The amount is not the issue. The issue is that there is a cost at all. "It's so cheap" is never an excuse for charging for something that should be free. In this case, running software you have no intent to charge for, on your computer. It's as if someone started charging $0.01/month for breathable air. "But $0.01 is trivial," would not excuse it.
It costs money, and isn't free, for a reason you're not acknowledging. I don't think it's a major profit center for Apple.
It's about setting a higher floor for malicious actors than "random botnet residential IP + a captcha solving service". It's about proving some semblance of identity through a card number and a transaction that goes through without a chargeback.
As the case upthread shows, there's plenty to dislike about a system that inhibits running code built for personal use. And it's obviously neither foolproof nor without collateral damage. Reasonable people can debate if it's worth it. But it still ought be acknowledged that the motivations are closer to the reason you have to identify yourself and pay a nominal fee to drive a vehicle on public roads.
I don't buy it. Or rather, I am willing to believe that some team at Apple has convinced itself that this makes sense, but they're wrong.
In particular, the security boundaries are nonsensical. The whole model of "notarization" is that the developer of some software has convinced Apple that the software as a whole (not a specific running instance) is worthy of doing a specific thing to the system as a whole.
But this is almost useless. Should Facebook be allowed to do various things that can violate privacy and steal data? What if the app has a valid reason to sometimes do those things?
Or, more egregiously, consider something like VSCode. I run it, and the fancy Apple sandbox helpfully asks me if I want to grant access to "Documents." The answer is really "no! -- I want to grant access to the specific folders that I want this workspace to access", but MacOS isn't even close to being able to understand that. So instead, one needs to grant permission, at which point, the user is completely pwned, as VSCode is wildly insecure.
So no, I really don't believe that MacOS's security model makes its users meaningfully more secure. At best, the code signing scheme has some value for attribution after an attack occurs, but most attacks seem to involve stolen credentials, and I bet a bunch just hijack validly-notarized-but-insecure software a la the VSCode example.
Notarization is not a trusted system on macOS - or rather, notarized binaries still have a "this was downloaded from the internet" prompt, and the user is meant to make a decision on whether it is trustworthy.
Notarization does some minimal checks, but is mostly about attaching a real identity so that maliciousness has at least some real-world consequences. The most obvious being that you lose the ability to get more apps notarized.
Actually the cost is not the issue (you are paying for it one way or the other), the issue is the authorization to do such an action on your (supposedly) own hardware.
The tool is built deep in our CI/CD chain. The whole thing is a house of cards built on a massive pile of tinder next to an open drum of kerosene. You want me to integrate XCode into that?
Last time I tried setting up an Apple developer license inside a large corporation, one that they paid for and not tied to me or my credit card, it was also a nightmare.
Who said anything about Xcode? The codesign tool is part of macOS, not Xcode. The CLI tool for notarization is bundled with Xcode, but you don't have to use it; they have an official REST API that you can use directly.
Sure it's trivial, but it is tacit acceptance that you need permission to make a program on their platform. Permission that needs to be renewed year over year. Permission to earn a living on this platform.
Permission that can be revoked for any reason, including being compelled by someone with more power than Apple.
...and there's an apple developer support person, Quinn, who appears to be heavily if not solely dedicated to helping developers do binary signing/notarization/stapling correctly.
Quinn also has their email address in their sig so people can just reach out via email without even needing an Apple account, or if they prefer more confidentiality.
As someone who actually signs, notorizes and distributes desktop apps for macOS, I can safely say their documentation is less than ideal.
Maybe because I'm using Electron framework which makes things more complicated, but I don't really understand why there's is a difference between different types of certificates (Developer ID, Apple distribution, macOS distribution) and I had to guess which one to use everytime I set it up.
Also why is notorization a completely different process from code signing, and requires completely different set of credentials from it. Seems odd to me.
> Also why is notorization a completely different process from code signing
Because they do completely different things. Signing is a proof that you were the one to write and package that software; notarisation is an online security check for malware. If I recall, you still sign but do not notarise when distributing to the Mac App Store.
OMG, this. I was working on a tool to help integrate password managers on macOS and I got completely blocked by the notarizing requirements. Some things literally cannot be built for macOS as open source software, now.
I don't really think saying documentation exists says much when Apple is notorious for having documentation that's either borderline or downright useless. It's generally the norm that some random blog post from a decade ago is more useful than their documentation, and I say this from firsthand experience.
Can you sign and notarize your own software made for internal use with your own infrastructure? If so, then this is a valid response. If not, then this is an irrelevant response because the issue is going through Apple, not the process being difficult or undocumented. If I own the device, then I should be free to decide what the sources of authority over it are.
Edit: I haven't tested it yet, but it does seem that you can sign an executable with your own certificate (self-signed or internal CA-issued) however you can't notarize it. Right now, notarization is only required for certain kinds of Apple-issued developer certificates, but that may change in the future.
Anecdotally, I was not able to find any way to notarize software for internal use, without paying for a $99 developer account. Though I would have been willing to pay, I know that others who might want to build the software wouldn’t, so I abandoned my project. I suppose I could have maintained it as open source with the developer account required to build, but it seemed disingenuous to me at the time.
> I mean, come on.
Is that really necessary? Obviously there are enough people who did not know about, or find helpful, the resources you’re referring to, that we have people complaining on Hacker News. This isn’t exactly a novice’s forum. Perhaps the problem lies with visibility and accessibility of the support resources, rather than all of the people who have seen notarization as a hurdle to getting real work done.
btw, for those who don’t want to search, Quinn’s signature states:
“
Quinn “The Eskimo!” @ Developer Technical Support @ Apple
let myEmail = "eskimo" + "1" + "@" + "apple.com"
I once really urgently needed `nmap` to do some production debugging ASAP. Unfortunately, the security tools would flag this immediately on my machine, as I knew this from previous experiments. Solution - compile my own binary from sources, then quickly rename it. I assume that this "workaround" was totally fine for sec department. At least production got fixed and money kept flowing.
> At least production got fixed and money kept flowing.
You were denied the tools to get your job done. You've put yourself at risk by applying an unapproved workaround.
Never ever do this (unless you hold substantial shares). Let the company's bottom line take the fall. If that's the only thing they care about, that's your only way to make the problem visible.
Unfortunately the real world isn't black and white. Yes, according to the company policies, I should watch the world burn and do nothing, while looking at the company bleeding money due to customers SLA being broken. Of course, after submitting a ticket to get nmap approved, which takes days. Extra points if I'm on oncall, then racking that sweet incident money is great.
But the underlying SRE culture here is that, if you know what you are doing and have a functioning brain of a responsible person, you'd be forgiven a jump over the fence, if it means putting out a fire on the other side of it. We aren't kids.
There’s a middle ground. Get the appropriate stakeholders involved in the decision, including security. Let security be the ones to keep the system down, if it cones to that. Or, let the business operations folks make the decision to go over security’s head. Either way, this is not something an engineer tasked with fixing an outage should be making the decision on.
Engineers _should_ have leeway in how they resolve issues. As I read, though, you have a company policy which explicitly disallows the action you needed to take to fix the problem (if I misread, my apologies). Getting the stakeholders involved is the responsible thing to do when policies need to be broken.
Ideally, the way this kind of situation gets handled should be documented as part of a break-glass policy, so there’s no ambiguity. If that’s not the case, though, the business should get to decide, alongside the policy maker (e.g.: security), whether that policy should be broken as part of an emergency fix, and how to remediate the policy drift after the crisis.
If you’re all tight enough that you’re allowed to make these kinds of decisions in the heat of the moment, that’s great, but it should be agreed upon, and documented, beforehand.
Well I found out the hard way that company culture or values can mean nothing if you don't CYA. Granted, the shop was small enough that our team was in charge of both the security policies and ops, but still, on one unfortunate occasion I've stepped outside my area of responsibility to "do what's right" and got punished. The next time I've been in a similar situation - well, I've walked away from the fire and grabbed the popcorn.
By the way, I'm still burnt out. This work is stressful. Don't let it take away what's already scarce for you.
xattr -cr <file> should clear the "download" extended attribute and make it as if the software was compiled on the machine itself, bypassing the ever-so-annoying Gatekeeper.
For binary patching: codesign --force --deep -s - <file> (no developer ID required, "ad-hoc signing" is just updating a few hashes here and there). Note that you should otherwise not use codesign as it is the job of the linker to do it.
Very aware of the attributes, unfortunately these machines are on a global corporate network so there are layers and layers of monitoring software to prevent internal and external attacks. Changing perm bits on an OSX executable is instantly noted and sent upwards as a possible security breach.
Last time we did this I had to spend a week explaining to management that Macs could actually run software other than PowerPoint and it was necessary for our job.
The local workaround that we use is to just spin up a Linux VM and program devices from there. The less legal workaround is using WebUSB and I'm afraid to even tell the necessary people how I did it, because it's sitting out on a public-facing server.
I don’t know if it’s just me, but i want more Gatekeeper, not less - help me stay safer. Or is it a security theatre? Malware producers can sign things just fine?
The more Gatekeeper, the more used people get to clicking OK without considering what it means. No amount of software can prevent the social engineering of an actual malware that tells the user to just click that OK button that they already have to do on a regular basis. Less is more here. It's why Windows tuned down their UAC after Vista.
Predefined value on current macOS's Gatekeeper is "move to Bin" instead of OK. Other option is Done - which cancels opening action. If you want to bypass that, you need to go to system settings > privacy & security and manually allow particular app there.
It is not a consent prompt. You get a choice on whether to trash the binary or quit.
To run a non-motorized app requires you to open a separate app, navigate to the security section and select that you want to authorize the app to run.
Apple does not have any desire to make distribution of non-notarized binaries commercially viable.
And we've seen this change across all browsers. There no longer is a "continue" prompt for TLS issues. The result is, way fewer maintained sites go months with an expired certificate.
The hashes are completely anonymized and not that intrusive. I'd rather they do it that way and have a global view of possible malware attacks than the complete free-for-all that other platforms "enjoy".
But here's my (unpopular) take as a GNOME user and using Fedora immutable distros + flatpaks -- I suspect Linux is going to go in a broadly similar direction. Maybe not soon (even flatpaks aren't universally acclaimed), but sometime.
It doesn't matter whether it is anonymized. Apple has no business collecting information about what executables I am running on my own computer, or even whether I'm running executables at all. I don't care what their stated purpose is. I don't care what they want a "global view" of. It's my computer, not theirs.
I don't even mind that they've introduced a level on the totem pole that's above root. But on my computer, -I- should be the one at that level, not Apple.
to downvoters: you can think its not fair that apple effectively holds control of your device, true, but the only way you can change things is to not buy the products. If you buy it, you accept how it is. vote with your wallets, not in some internet forum
I think it depends on what distro you're talking about. Corporate distros like RHEL and SLES are absolutely going that way. It takes a lot of effort to backport fixes, and the money's not there in desktop Linux to make it worth their while if containerization is a viable alternative. Red Hat's gotten rid of a bunch of graphical applications for RHEL 10 and stated that users can get them from Flathub as an alternative. I believe there was some consternation when CentOS Stream 10 launched without even a packaged web browser and the advice was to install Firefox from Flathub (there's a lot of use cases where that breaks stuff), but it appears they've walked that back and started providing Firefox as a traditional package.
However, less corporate distros that mostly just ship built upstream software as-is since they don't have to support it for long periods (think Arch, Fedora, Void, etc) don't have that problem, so I expect we'll continue seeing them use traditional packages.
> I believe there was some consternation when CentOS Stream 10 launched without even a packaged web browser and the advice was to install Firefox from Flathub
Ubuntu does the exact same thing with their snap repository, the Firefox apt package from Ubuntu is fake. At least Flatpak is a community-led project unlike snap.
You can limit the file system permissions of the app, like giving only access to downloads, so that if/when there’s a sandbox leak you’re fine. You can also disable various things, like webcam or mic, this way.
In addition, you can get perpetual updates to the latest version of your browser even on old, stable distros like Debian.
Running a new browser on an old distro would be a strong reason for me (if I somehow couldn't update the distro - but I can and I do.)
Regarding security, the added work and complication outweighs the added security for me. I can't really disagree with having a different preference. More security on this wild internet is better, right?
IMO it's not much added work. In KDE you can navigate to settings and edit flatpak permissions, and flatpaks are available to download via discover. I haven't noticed any weirdness for firefox or chrome.
I understand and appreciate the sentiment, but I see the intent very differently. Apple is not employing a frog boiling strategy, but rather being responsive to an increasingly sophisticated adversary.
It’s like criticism of the quality of Google search dropping. It has absolutely tanked, but it’s not because the algorithm is worse, it’s because the internet has grown orders of magnitude and most of it uses the same hyper aggressive SEO optimisation, such that the signal to noise ratio is far worse than ever before.
You can also block specific subdomains, too. Useful when I want to be able to see finance.yahoo.com items in my search results, but nothing else from the yahoo.com domain.
That rationalization ignores a lot of confounding evidence, such as other search engines being able to deliver great results and adequately keep the SEO garbage out.
That’s kinda the SEO equivalent of security by obscurity though, right? SEO spam puts a lot less effort into optimizing for other search engines, whereas Google is dealing with being the primary target of every adversarial SEO spam site.
This is a great theory but it isn't the reason. Google management made a conscious decision about five years ago to prioritise profit over search quality.
We know this because the emails came out in discovery for one of the antitrust suits.
The biggest struggle is that the original Macintosh was so simple to manage. The original concept of system extensions to expand the capabilities and the file structure built on the hierarchy with the desktop as the top level was broken with the shift to Unix.
Suddenly the users file hierarchy started wherever the Home folder was located and it became an island of user controlled environment surrounded by complexity of computer operating systems.
The result I found overall well thought out but when the desktop became just a folder I felt the Mac moved from it’s simplicity embracing the complexity that was offered by windows.
Simplicity is fine for a hobby project. An operating system having zero concern for any kind of security is a non-starter today.
It's amazing the rose tinted glasses people have about the original Macintosh environment. It was insanely janky and (unless you were ruthlessly conservative) insanely unstable by today's standards. By version 10.5 (Leopard) the modern UNIX-based MacOS was unequivocally superior to Classic MacOS in every metric other than nostalgia.
I understand the trade offs and accept them. I was trying to point out where the split is and how it won't go back. I think the point of view expressed in your comment is s distorted as the ones your derding.
I also believe that the simplicity could have security as performant. The real advantage of the Unix layer is compatibility that the Macintosh was missing.
I sincerely tried to interpret what you meant here, but I failed. I understand the words, and the fragments of every sentence, but I wasn’t able to deduce the intent of your reply.
Are you trying to say that it’s possible for a system to be both simple and secure? Absolutely that’s the case, but with a trade-off — either it needs to restrict the user’s freedom, or be fully disconnected from the outside world.
I have pondering these ideas a long time and what is needed is an intense glossing over of all the details. The original Macintosh did exactly this and was called a toy and with 128k completely useless. Alternatively my unsophisticated Mom saw the Mac 128k demoed at the mall and went into a frenzy to get that tool. She wanted to publish documents.
The threats in the world are real and the internet doesn't help. I 100% agree that a network connection needs to be kept at a distance to make things simpler.
I think the power of language used to describe a system is where simplicity begins.
What I'm working on is creating a crisp line of delineation between "local" and "public" networks.
If by default after is on the "local" network auto-discovery is secure. If things are explicitly needed a user can publish them through physical manipulation to publish to the outside world.
The outside world can now be described using classic Users and Groups which is cultural easy to understand.
I'm trying to create an environment that focuses on making those 2 things plus a third element simple to understand and physically manipulatable.
The freedom I'm looking for is available on the "local" network. The "public" network is where our data is interchanged with the outside world based on our publishing. I don't expect people to interact with this layer much. I expect people to configure it for whatever institution/organization/government.
Most of the complexity I see in computing these days is market drive demand for eyeballs/clicks/...
Of course Google's algorithm is worse. Google prioritises showing you search results that make money for Google. Google has no incentive to show you anything else.
I can't believe I even have to say this out loud. Look up enshittification.
Actively depleting the good-will they accumulated over the years definitely makes it worse. It's that harder to give the benefit of the doubt to a company also showing the middle finger to their Devs.
> Google
Giving priority to AdSense sites, fucking around with content lengths (famously penalising short stay sites), killing advanced search options. That's just thinking about it for 10s, but to me most of it is totally of Google's making.
The difference: in Linux it is a usability issue to be fixed, whereas on macOS it is a feature and explicit design goal to make it that way. In general, I have found that things which are difficult on Linux are so because the problem is difficult, not because the people who make my computer have paternalistic attitudes about my usage of it.
So it’s purely ideological without any real world difference?
Are most people better off with Apple defaults?
And it’s not because the problem is “difficult”. It’s because for 20 years it has been claimed that this will be the “year of Linux on the Desktop” and it’s never been good enough for most people.
That isn't the denotation of my post; I was not characterizing Linux as a whole, but only responding to your specific (unsound) analogy. It works better for me, for a number of reasons including that above. Perhaps it will work better for you, as well. :)
The second part of your post is incoherent to me, I can't tell what you're trying to say.
It’s perfectly fine. KDE and Gnome are both now more cohesive, more intuitive, and less buggy than either Windows or MacOS.
The problem with Linux is that, while it’s very good, it’s different.
Nobody actually cares how intuitive something is, at least not in absolute. People will still say Windows is intuitive. Pretty much nothing in Windows, from the registry to COM to IIS to setting/control panel/computer management, is intuitive. But they know how to use it and are used to that particular brand of buggy inconsistency.
Linux desktops have been high quality for a long time now. The reality is you, and others, measure quality as “how much is it like windows” or “how much of it is like macOS”. When that’s your metric, Linux will always come up short, just by definition.
If I pick up a Linux laptop right now, how well will it handle low latency audio? How well will it handle power management? My graphics hardware? Getting WiFi to work reliably is still an issue.
Can I get a high performance Linux laptop with good battery life, fast graphics, that runs cool and silent?
I can certainly get a Framework (Fedora and Ubuntu officially supported), throw my prefered Bluefin-Framework image in and get working
Battery life around 7 hours is the average I see reported, Fast/Silent will depend on the model, but I don't see the issue really
Upgradability and easeness of battery replacement are a plus
I just picked framework because they were first to come to mind, but I think Dell has a nice Linux story, Tuxedo also comes to mind
7 hours battery life is less than half of what I get on my MacBook Air. That wouldn’t last me on my ATL - HNL flight I took last year or my MCO - LHR 10 hour flight I’m taking this year.
These are the typical reviews I see around the Framework
Yes, I just bought one a few months ago actually. A new lunar lake laptop. It gets 12 hours of battery life and has plenty performance for programming, plus 32 gigs of ram. It’s under 3 pounds and the screen is OLED.
And yes, everything works. On bleeding edge 2 month old hardware.
I even use thunderbolt 4 to connect my external displays and peripherals. Not only does it work, but it’s pleasant. KDE has a settings panel for thunderbolt. I can even change my monitor brightness in KDE settings. No OSD required!
But wait, there’s more! I’m running 2 1440p monitors at 240hz and the system never even hiccups.
But wait, there’s more more! The battery settings are really advanced so I can change the power profile, maximum charge, everything.
The only thing I’m unsure about in your comment is “low latency audio”. It seems low latency to me, but I’m not an audio engineer.
> Getting WiFi to work reliably is still an issue.
This should not be an issue. I have hardware that varies a lot and I literally buy random wifi dongles for $1, $4, $5, Amazon, AliExpress, etc. and they have all just worked on first plugin. I can easily take my phone and tether it to my PC using USB-C and it appears in my Gnome network list and just starts using it for Internet.
> how well will it handle low latency audio
Pretty well you can use OBS to verify this. There are plenty of settings if you want to tune that.
> My graphics hardware?
Just ignore Nvidia and move on. Sure they might figure it out one day, I gave up a decade ago and I use Intel integrated or AMD dedicated for GPUs. Nvidia does "work" for most purposes but it will cause you a headache eventually and those are not worth $400 to me.
> How well will it handle power management?
I enjoy the basic controls that Gnome provides that give me a simple way to basically say "go all out" or "save some battery" etc. There are finer grain controls available and I have used commands in the past to push crappy hardware to it's limits before I chucked it (old Intel iGPUs)
> Can I get a high performance Linux laptop with good battery life, fast graphics, that runs cool and silent?
You can get ones that are specifically marketed for this purpose. Tuxedo is one that specializes in this and obviously System76 also do. These have a higher price point than a regular Dell system, which IMO is the better option in some ways. Dell sells more systems and has more users and it will "just work". They sold Linux systems for years and still do I believe.
Regarding "running silent" this is a gripe I have, not that it runs loud but some laptops have custom RGB crap and sometimes in Linux I don't have access to the extra functionality to customize my lighting or manually set my fans to ramp up etc. There are projects that aim to do this, but I have not looked into them beyond the most basic `fancontrol` built in command.
I think once you expand the scope to "most people" it might become impossible to say what the correct answer for that large of a group is. In the past their value add might have been more compelling and their feature lock not as draconian. It appears some people think that has changed over time.
> In general, I have found that things which are difficult on Linux are so because the problem is difficult [...].
Hard disagree. Audio mixing is not difficult[1]. The Linux kernel guys were right - it does not belong in the kernel. The userspace story however, has been a complete shitshow for decades. I think Pipewire mostly fixed that? Not sure, sometimes I still have to log out and back in to fix audio.
The funniest part? It's been working in the BSDs all along. I recommend reading the source of sndiod[1].
What's even worse? Probably systemd. I try not to hold a strong opinion - I tolerate it, the way I tolerate traffic when taking a walk. The technical issue however is several orders of magnitude simpler - again, the BSDs come to mind, but you can also write a complete PID1 program in about 20 lines of straightforward C[2]. I don't mind the commands being unfamiliar (they're already all different in almost every OS family); it's that the entire package is dreadfully large in scope, opaque, and I find it more difficult to interact with than anything else in this landscape.
There are indeed always exceptions to generalizations, as you've pointed out. Though pulseaudio always trudged along fine for me (not like audio had always worked for me on other systems), and pipewire works perfectly.
I agree PulseAudio, Pipewire, ALSA, etc. are a pretty big shit show in Linux and have been for some time. From what I understand there are a few stories there with various levels of screw ups, but at no point was this situation the goal, and we are moving closer to an easy to use system that "just works" for these needs.
However, it's worth noting that audio experts doing high grade mixing in production are using these systems quite effectively and have been for a long time. It's similar to Blender in that regard with it always having the "guts" of doing great things, but only the experts that knew the correct spells to cast were able to use it effectively before the UI/UX was improved with 2.x and later I believe.
I work in live media production. I would never consider doing any mixing on Linux - just like I wouldn't consider putting Docker containers on a Mac to serve live HTTP traffic.
Don't disable SIP, clear the downloaded/quarantine extended attribute instead. This clears all extended attributes: xattr -cr <file> and bypasses the obnoxious GK.
Frustrating thing is the earlier versions worked well, it protected you from accidental things but the way to force it was clear and obvious. Now bypass is obtuse and requires enough work arounds people advise just disabling it which is also bad to normalize.
I migrated to Linux about a year ago too. Not the smoothest experience ever (looking at you, ath11k with device-specific quirks) but so far I am delighted. Finally, I don't have to fight my computer to do things I expect it to do.
Unfortunately, I still have to deal with macOS for work due to corporate policies.
The main problem I had with living in a Gnome desktop environment, is with the keyboard. I'm not willing to abandon my use of Emacs control+meta sequences for cursor and editing movements everywhere in the GUI. On macOS, this works because the command (super/Win on Linux/Windows) key is used for common shortcuts and the control key is free for editing shortcuts.
I spent a day or so hacking around with kanata[0], which is a kernel level keyboard remapping tool, that lets you define keyboard mapping layers in a similar way you might with QMK firmware. When I press the 'super/win/cmd' it activates a layer which maps certain sequences to their control equivalents, so I can create tabs, close windows, copy and paste (and many more) like my macOS muscle memory wants to do. Other super key sequences (like Super-L for lock desktop or Super-Tab for window cycling) are unchanged. Furthermore, when I hit the control or meta/alt/option key, it activates a layer where Emacs editing keys are emulated using the Gnome equivalents. For example, C-a and C-e are mapped to home/end, etc.
The only problem is, this is not the behavior I want in terminals or in GNU/Emacs itself. So I installed a Gnome shell extension[1] that exports information about the active window state to a DBUS endpoint. That let me write a small python daemon (managed by a systemd user service) which wakes up whenever the active window changes. Based on this info, I send a message to the TCP server that kanata (also managed by a systemd user service) provides for remote control to switch to the appropriate layer.
After doing this, and tweaking my Gnome setup for another day or so, I am just as comfortable on my Linux machine as I was on my Mac. My main applications are Emacs, Firefox, Mattermost, Slack, ChatGPT, Discord, Kitty, and Steam. My Linux box was previously my Windows gaming box (don't get me started about frog boiling on Windows) and I'm amazed that I can play all my favorite titles (Manor Lords, Hell Let Loose, Foundation, Arma Reforager) on Linux with Proton.
I'm convinced a DE that figures this shit out out of the box will explode in popularity. Super for the OS and DE shortcuts. Ctrl for the Terminal and readline cursor movements. It can't be impossible to bake these in as defaults.
Love this, and I'm in the same boat. Is your configuration of kanata public at all?
I know it's mostly muscle memory, but macOS shortcuts just seem sane and consistent and that has been one of the biggest frustrations when trying to switch. I found toshy[0] which does something similar - did you try that? The goal is purely macOS key remappings in Linux, so a much smaller scope than kanata.
I didn't try toshy, I had a bad experience when I tried kinto.sh a couple of years back, and I had a pretty clear idea of how I could get what I wanted out of a fully featured keyboard remapping tool under Linux. I initially started with Kmonad, but once I found Kanata, and realized that it had a TCP interface for programmatically changing layers, I quickly switched.
I have a Kinesis 360 keyboard, and my config[0] probably won't work for other keyboards, but it can give you a starting point for your own config.
One thing that has been slowly creeping in is a little bit of a Microsoft-like "you will use our feature", like launching apple music every time I hit headphone controls, or nagging me to turn on reactions every time I start a video call. In some ways that's more annoying than the outright bugs, as they could choose not to be that way and market themselves as not being that way.
I feel your pain. I hate pushy upsells and promos. Also the cluttered settings App "Remember to setup Apple Pay" promos. I do value user education. They need to consolidate all of the feature promo services into a revised Tips tool that allow users to engage with new features at their own pace.
Agree. Apple needs to clean up shop - MacOS has been egregiously worsening year over year. Some features like Universal Control and Continuity Camera are legitimately awesome, but they do not make up for the INSANELY slow System Settings app that gets harder to navigate with each release and which has >2s wait times for the right pane to respond to a change in the left pane. Steve Jobs would have fired the person responsible for that overhaul three years ago, it's embarrassing. Messages too needs a ground-up rewrite. Getting more elaborate emoji tapbacks doesn't make up for fundamental instability and poor syncing behavior. C'mon!
Absolutely. I love the work they have been doing on the backend, like PQ3 [1], but it just doesn't work for me when the Stickers and Emojis extensions on Mac leak several GBs of RAM and I have to terminate it several times a day to free up memory.
Another thing I dislike is that it stores the whole message history on the device. It's nice to have at times, but I send a lot of photos, which adds up in storage over time. I pay for iCloud, and store my messages there. Why does my Mac need to hold every single photo I have ever sent?
Local iMessage storage is debilitating. I have over 90GB of iMessage history that I don't want deleted. The keep messages for x days removes it from iCloud and the Mac though. Why?
System Settings is awful. Whoever decided to hide tons of settings inside innocuous "(i)" non-buttons should be kept far away from UX design. It's the hamburger menu of macOS.
> Getting more elaborate emoji tapbacks doesn't make up for fundamental instability and poor syncing behavior. C'mon!
Oh but you forgot about the “catch up” button they added 2 releases ago that takes you to the last unread message! …
… but only if said last message is within the N most recent messages, in the messages which are already “fetched” from local storage. If it’s more unread messages than that, the button is nowhere to be found.
Like they said “ok we can implement a catch up button but it’ll be hard to solve due to how we do paging.” “Ok we just won’t put the button on screen if we have to page then. Save the hard problem for the next release.” Then they just forgot about it.
I'm enjoying Sorbet Leopard on my 20 year old Dual Core PowerPC tower. Mostly just messing around with old versions of Max making weird sounds.. but when I do interact with the OS it feels great and responsive and a joy to use. Modern MacOS can feel that way if you turn off a lot of crap. I don't even sync my accounts to the OS anymore.
- Ever since I've updated to the latest iOS 18, my watch complications(weather doodad) stop working randomly because they just lose the location services permission. Then in settings, the location services permission list acts like the weather app isn't installed.
- The new Mail app now automatically classifies your email, but still gives you the "All Mail" option. But the unread count badge on the app only works off of what they classify as your "Priority" mail. There's a setting to change that, so that it shows you the unread count of ALL mail, not just priority mail, but when you change that setting nothing changes. This is my biggest problem with new iOS.
- Keyboard sometimes doesn't get out the way any more when it should.
These are just off the top of my head. It used to be such a nice, polished experience. Their competition was just outclassed. Now, when my phone dies I'm going to have a good look at all the other options.
> - Keyboard sometimes doesn't get out the way any more when it should.
Depends on where you were seeing this of course, but this could very well be an app problem instead of a system problem.
Native UIKit/SwiftUI do a little bit of keyboard management for “free”, but there are many circumstances where it falls on the developer’s shoulders to do this. For cross platform frameworks, some do keyboard management others don’t even try. For web apps it’s a coin toss and depends on which of the gazillion ways the dev built their app.
It’s not actually that hard, usually just a matter of making sure that your scrolling content either resizes to match the keyboard-shrunken viewport or adding bottom padding equivalent to the height of the keyboard and then and adjusting scroll position accordingly, but it’s not unusual to see this partially or fully absent, especially on poorly built cheapest-bidder-contracted apps.
In modern UIKit it's as simple as constraining to the keyboard layout guide. That gives you full animation support for free as well, no more need to listen for the notification and manually set up animations with the same timing and curve. On iPads the keyboard guide can even help you avoid the split keyboard, it's really nice.
Of course SwiftUI gives you almost none of this control, forcing you to hope the magic automatic support works how you expect.
But then neither help you with any of the other interactions, like any background dimming you may want, or tapping away from the keyboard to dismiss. That has to be done manually.
my iPhone gets into a state lately where a pane will suddenly lose the the ability to _scroll_. it can happen in any app, but I see it a lot in Safari. Like, what is even happening, this is a fundamental UI interaction. The only way to fix it is to close the tab or force-quit the app. Super weird.
Absolutely. And turning off Siri's "Learn from this app" should not require the user to navigate to every single app's menu, when Siri has a top level page in Settings.
The division of per-app vs app list in general is bad.
I think they should just throw in the towel and duplicate settings. Meaning, we can turn off Siri learning from an app or from the Siri page. Or we can turn off banners from the app or the notifications page.
I don't think it does. Long term Apple user here (since 2007). I'm typing this on a 5 year old pile of junk with Windows 11 LTSC on it. The (M4) Mac is sitting next to me acting as an SMB server until I can be bothered to get all my stuff out of it. It's just tiring using a Mac these days. It's difficult to explain but everything feels slightly frustrating. The nice things are really nice. The whole experience is quite nice. Until you hit a problem. Then it's a complete pit of pain and misery and there just aren't enough ways out of it.
Had a few issues with iCloud syncing and data loss as well and what with being based in the UK and the general problems with geopolitics and the cloud I figured I'd try and get as much stuff out of iCloud as possible. Well there's not much advantage now. Most of it is in the ecosystem tie in, not the hardware. And on top of that the provisioned services such as Apple Music are just pain for me on a daily basis. My entire music catalogue disappeared in a puff of smoke when I was offline for nearly a week. The one thing I wanted it for!
So back to the PC. I ran out of disk space on the (soldered in SSD) Mac. I can't delete anything and macOS has leaked out about 20gb suddenly. I don't know what this is other than about 5 gig of it is Apple Intelligence despite telling it to fuck off. So it's late Friday afternoon and I need to get something done so I can have a clear weekend. I dig in the junk cupboard and find a couple of hard disks but no way of connecting them to the USB-C only Mac. Amazon solutions aren't available for delivery until Sunday. There upon I discovered the kids' "covid work PC" for when they were home studying. Despite the acceptable 16Gb of RAM it only had a meagre 256Gb disk in it. No worries. Opened it up and there's a hole for an SSD in it. It now has +500Gb SSD. Brilliant. On goes windows 11 LTSC. I'm back up running R in under an hour and have transferred all the data over.
I never went back. It feels better here. This thing is a swiss army knife. And extension of me. Not the other way round like on the Mac. The Mac feels like it feeds off me: both cash and energy. Apple need to fix that.
That would be medium-term user. Long-term would be people like me that have been using it since 1984.
I have a collection of Macs going all the way back to 1984. Even the newest one hasn't been turned on in three years.
My daily driver is Windows Server 2016. But it has VMware Workstation so there are lots of virtual machines for my work, including Linux.
I am so tempted by the new M4s. Amazing piece of technology. So sad about the operating system though. Every year I say I'll wait for a quality Linux port.
what are some good alternatives to mac os? there some features like image/text copy-paste being cross device that are insanely useful that make it hard to switch
There are three main choices and they are all compromised in their own way. You just need to figure out what is important to you and what isn't.
What you shouldn't do is take too much notice of posts like these, I've read through the whole thing and haven't had any of the issues mentioned. I've also not seen a mention of the issues I do have. HN has a negative tone, it seems we like to whinge.
> I could walk item by item through System Settings and point out many equally inexplicable decisions. Did anyone at Apple really believe a Mac user’s life would be better if common features were buried deep in menus?
I have to agree with this, System Settings seems very inconsistent (design) and has terrible information architecture / organization.
799 comments
[ 3.3 ms ] story [ 523 ms ] threadThere is no silver bullet, just a lot of lead ones and the answer to Apple's quality problem is to begin baking QA back into the process in a meaningful way after letting it atrophy for the last decade or so.
Hire more humans and rely less on automation. Trust your developers, QA, and user support folks and the feedback they push up the chain of command. Fix bugs as the arise instead of assigning them to "future" or whatever. Don't release features until they're sufficient stable.
This is all basic stuff for a software company, stuff that Apple seems to have forgotten under the leadership of that glorified accountant, Cook.
In practice it's a challenge because the OS bundles a lot of separate things into releases, namely Safari changes are tied to OS changes which are tied to Apple Pay features which are tied to so on and so on.
It would require a lot of feature flagging and extra complexity which may reduce complexity.
Another way is to start un-bundling releases and fundamentally re-thinking how the dependency graph is structured.
But, in a sense, this still incorporates your idea, because the devs and QA must be given the mandate of finding these bugs, and also towards making the automated tests cover the bug's related test cases (as well as charged with improving the test code itself, which is often in a mediocre state in most code bases I've seen at least).
By all means people should complain on forums (why not?), but a forum post complaining about some years-old bug isn't going to be anywhere near as effective as contacting apple's support or filing a bug report.
I'm not a developer, I'm just a regular user - so if I can get all this special treatment, so can you.
(It feels similar to how those same podcasters absolutely blast Apple Intelligence, while non-tech users I've heard from seem to love it.)
As a former Apple employee of 13 years: Apple knows about the bugs. QA isn’t the problem.
A lot of people complain that their radar for some obvious bug isn’t getting noticed, and conclude that Apple must not be QA’ing, or not dogfooding their own product. This isn’t the case at all. I guarantee the bugs you care about are well known, and QA has already spotted them.
The reality is, they just don’t care. The train leaves the station in September. You’re either on it or you’re not. If you spent the year rewriting some subsystem, and it’s July and you have this huge list of bugs, there’s a go/no-go decision, and the answer is nearly always “go” (because no-go would mean reverting a ton of other stuff too, and that carries its own regression risk, etc.)
So instead there’s just an amount of bugginess that’s deemed acceptable. And so the software is released, everybody slaps high-fives, and the remaining bugs are punted to next year, where they will sit forever, because once we do one release with a known bug, it couldn’t be that important, right? After all, we shipped with it! Future/P2, never to be seen again.
An attempt was made to remedy this by pushing deadlines earlier in the cycle, to make room for more QA time, but that just introduced more perverse incentives: people started landing big features in later dot-releases where there’s less scrutiny, and even more tolerance for bugs.
The honest answer is that Apple needs to start giving a damn about the quality of what they’re pushing. As Steve once said at a pretty famous internal meeting, “you should be mad at your teammates for letting each other down like this”. And heads need to roll. I can only hope that they’re realizing this now, but I don’t feel like the culture under Tim works this way. People’s feelings are way too important, and necessary changes don't get made.
I miss those. Unfortunately, since Apple doesn't do the whole space theme anymore, you'd probably get some really boring drone shots of California at best before a Setup Assistant faded into view from behind a Redwood or something.
I had to hear those goddamn songs so many times, often all at the same time.
I'm weird though, and never stopped liking it.
Doo-do-doo-doo-do...
Indeed, I remember three times when Apple went a bit overboard on the feature front, but dialed it back and made some of the most stable and useful OS versions:
OS 8.5/8.6 pushed a bunch of features and were the last big pushes pre-OSX, but then OS 9 fixed a TON of bugs, and added a few smaller quality of life improvements that made running 'Classic' Mac OS pretty good, for those who were stuck on it for the transitional years.
Mac OS X 10.0 rewrote _everything_, and especially 10.0 was _dog_ slow, with all the new Quartz graphics stuff in an era where GPU accelerated 3D display widgets wasn't quite prevalent. 10.1 patched in a bunch of missing features (like DVD Player—it was still a pretty useful tool back then), and fixed a couple of the most glaring problems... but 10.4 Tiger was the first OS X release that was 'fast' enough OS X was a joy to use in the same way OS 9 was at the time. At least on newer Macs.
And then of course Snow Leopard, which is the subject of the OP.
macOS 13/14/15 have progressively added more little bugs I track in my https://github.com/geerlingguy/mac-dev-playbook project; anything from little networking bugs to weird preferences that can't be automated, or don't even work at all when you try toggling them.
That's besides the absolute _disaster_ that is modern System Preferences. Until the 'great iOSification' a few years back, Apple's System Preferences and preference pane were actually a pleasure to use, and I could usually remember where to go visually, with a nice search assistant.
Now... it's hit or miss if I can even find a setting :(
There was no acceleration (even 2D!) until 10.2
That said, I expect things to get worse as they manage to converge their multiple platforms in exactly the wrong way (by dumbing them down across the board even as people keep hoping they'll make iPad OS more useful, etc.).
But at least we still have Safari, Apple Silicon is pretty amazing and I can survive inside Terminal and vim. For now.
I entirely forgot it existed! They still sell that?
It's just that me and other old-time switchers have stopped complaining about it and moved on (taoofmac.com, my blog, was started when I wrote a few very popular switcher guides, and even though I kept using the same domain name I see myself as a UNIX guy, not "just" a Mac user).
For me, Spotlight is no longer (anywhere) near as useful to find files (and sometimes forgets app and shortcut names it found perfectly fine 5 minutes ago), and there is no longer any way to effectively prioritize the results I want (apps, not internet garbage).
Most of the other examples in the article also apply, but to be honest I've been using GNOME in parallel for years now and I consider it to be my "forever desktop" if PC hardware can ever match Apple Silicon (or, most likely, if I want something that is _just a computer_).
Settings -> Spotlight -> Websites, UNCHECK
Also, I vaguely remember there being a way to _order_ results, not just disable them.
Good memory! Apple removed this feature in El Capitan.
My C drive was was super full for some reason I couldn't understand, and Explorer couldn't tell me where the data was. There was about 100GB just unaccounted for.
I don't even use the search index.
Then when it does show the results, they’re usually in some terribly unhelpful order. It took me ages to try and go through the CUJ of “this app isn’t sending me notifications because I turned them off now I want them back on”
People don't want animojis, and they don't want other trite new features that only seem to exist because Apple feels it needs to demo something new every year.
What they want is something that just works without annoyances, distractions, failures, or complications.
Give them that and they'll break down the doors trying to get their hands on it, because it's so far from how most tech works today.
Yeah and they succeeded in that so now it's about selling subscriptions on top of that.
I'm there as well. I've been really enjoying desktop Linux lately, but I can't go back to a non-Apple laptop at this point. There's just nothing else on the market that comes close, they all make some tradeoff I'm not willing to make - either screen, speakers, keyboard, heat/battery life/fan noise, touchpad, etc. Apple is the only one that has the entire package.
There's Asahi, but no thunderbolt yet and I'm not sure the future of that project with the lead burning out and quitting. I just want an Apple Silicon-esque laptop, no trade offs on components, that runs Linux, and there's no OEM out there that's offering that experience.
So, until that happens I'm staying on mac, and even with declining quality, it's not all that bad compared to the alternatives yet. I've learned to mostly work around/ignore the odd bugs.
The real problem is that, just like the grandparent post pointed out, Apple's software quality has been declining. The Tiger to Snow Leopard epoch was incredible. Apps were simple, skeumorphic, and robust.
Right now, the whole system feels a lot less coherent and robustness has declined. IMHO, there are not so many extra features worth adding. They should focus on making all software robust and secure. Robustness should come from better languages that are safe by construction. Apple can afford to invest on this due to their vertical integration.
Because right now it’s clearly so far down, beneath dozens of other priorities, that expecting it to just happen one day seems futile.
Unfortunately:
1) AVP is about 10x too pricy for games.
2) It's not clear if it can beat even the cheapest headsets for anything important for telepresence (higher resolution isn't always important, but can be sometimes).
Irregardless, you need the associated telepresence robot, and despite the obvious name, the closest Apple gets to iRobot is if someone bought a vaccum cleaner because Apple doesn't even have the trademark.
3) (a) is creepy, and modern AI assistants are the SOTA for (b) and yet still only "neat" rather than actually achieving the AR vision since at least Microsoft's Hololens, and because AI assistants are free apps on your phone, they can't justify a €4k headset — someone would need a fantastic proprieraty AI breakthrough to justify it.
The things it does may not seem important today, but back then even just my bandwidth costs were a significant percentage of my shareware revenue.
ObjC with manual reference counting wasn't much fun either; while we can blame Apple for choosing ObjC in the first place, they definitely improved things.
Now that the platform is cemented, they don't have an incentive to cater to developers.
The reality is that there was a long period of time where Apple built up lots of goodwill with a developer ecosystem that exceeded by many orders of magnitude the pre-iPhone OS X indie Mac developer scene.
There were many, many developers that hadn’t even touched a Mac before the iPhone came out, and were happy with Apple, and now are certainly not.
Another way to see it is that people who programmed for Mac OS already had reasons to be annoyed by Apple (e.g. 64bit Carbon). The iPhone let it get new people, who eventually found out why the pre-iPhone scene felt that way.
Expecting users to change their daily habits in order to marginally improve the operating system of a trillion dollar company feels naive and a bit disrespectful to people who actually use these machines for work.
Even developers… the vast majority of developers ignored Apple for decades (and Apple was also hostile) and it managed to grow despite that.
Might as well ask people to contribute to Gnome or whatever so in the future everyone can go somewhere better. Feels way more feasible.
A sentiment which famously led Steve Jobs to respond that he doesn't understand this, because "people pay us to make that decision for them" and "If people like our products they will buy them; if they don't, they won't" [0]
So according to Steve Jobs himself, the only Apple-acknowledged way to disagree with Apple is to NOT buy their products, and by extend into the services-world of today it means STOP USING their products.
Now Steve Jobs doesn't officially run this company anymore, but I don't see any indication that this philosophy has changed in any way.
[0]: https://www.youtube.com/watch?v=i5f8bqYYwps&t=772s
Most people are not going to migrate to Android, Windows, Linux or whatever else just to make macOS marginally better.
And it's fine: marginal quality improvements of a product are not the "responsibility" of consumers.
Nobody is saying “gosh, macOS is so damned unstable, but I’ve gotta use it, because…blue bubbles on my iPhone?
You’ve just read some story about a company you already hate and are parroting it.
It's not just "blue bubbles," but "blue bubbles" seems like a good shorthand to me. It's also things like Hand-off, or Universal Control, or getting Messages on both iPhone and Mac seamlessly, or being on the same WiFi network allowing your iPhone/Watch to work as a tv remote for the Apple TV even if you're just visiting a friend. Features that any platform can and does enable, but that do to Apples vertical can work seamlessly out of the box, across all the product lines, while securing network access in the ways most users will want, creating a continuous buy-in loop wherein the more Apple products you buy, the more incentive there is to buy exclusively Apple.
And it's a collective "you." If your entire family uses exclusively Apple products, then you'll be the only person who can't easily use eg the Apple TV in the living room, or the person "messing up" the group chats with "User reacted with Emoji Heart to [3 paragraph text message]," or the one trying to decide between competing network KVM software platforms so that you can use your tablet when your 12-yo can just set their tablet next to their laptop and get a second screen without any setup. Nevermind that these are all social engineering techniques that only exist BECAUSE Apple chose not to play nice with others, they still socially reinforce a deeper commitment to Apple products with each additional Apple product in the ecosystem.
I say this as someone "stuck in the blue bubble" with eyes open about what's going on. I'll keep picking Apple as long as they're a hardware-oriented company, because their incentives are best aligned with mine for the consumer features they are delivering (for now): consumer integration that sells hardware. It's insidious in its own way, but not like "hardware that sells eyeballs" (Google/Meta) or "business integration that sells compliance" (Microsoft).
Regardless: that kind of message doesn't feel like HN-worthy productive discussion.
So you get rid of removable batteries so customers have to toss their phones away more often, you gimp other feature, you spend more money on advertising than you did actually developing the product (read this bit several times until it sinks in how crazy it is, yet that's how we are with every major phone, every major movie, etc), and so on.
I don't doubt that after 2020 the advertising budgets far outstripped the production budgets - multiple times; I am curious if that trend continues now, now that production isn't hamstrung by covid restrictions.
Capitalism works this way because its customers, the investors, want it to work this way, because growth is how you get compound interest. Investors include anyone with an interest bearing bank deposit, a 401k, stocks, bonds, etc.
No growth means it would no longer be possible for an investment to appreciate.
I think of a similar thing when I see people complaining about how companies don't want to pay good wages. When you go shopping do you buy the $10 product or the $5 essentially equivalent alternative? Most people will buy the $5 one. If you do that, you're putting downward pressure on wages.
It's in your (purely economic) best interest for your wages to be high but everyone else's to be low. That's because when you're a worker you are a seller of labor, while when you're a customer you are an (indirect) buyer of labor.
Everything in economics is like this. Everything is a paradox. Everything is a feedback loop. Every transaction has two parties, and in some cases you are both parties depending on what "hat" you are wearing at the moment.
Equity returns ultimately come from risk premiums. (Which are small now in US equities BTW).
I’m invested in a microcap private equity fund that has returned >20-25% for years. They have high returns because they buy firms at 3-4x cashflow. You will get the high returns even with no growth. And with no increase in valuation. The returns are a function of an illiquidity premium.
With Apple explicitly, growth is expected given the valuation level. If it doesn’t grow, the share price will decline. So yes, in their case, firm is certainly under pressure to grow.
I also don’t agree with your “best interest for wages to be high and everyone else’s lower”. That is one aspect. It is more complicated. Consider Baumol Effect for starters.
Things like retirement, 401ks, etc., are society-wide institutions subject to macroeconomic rules.
There are also entire "industries" designed to shield people who want to find quality content from big 'A' advertising.
https://www.youtube.com/watch?v=tGKsbt5wii0 For context John Sculley said "Apple was the marketing company of the decade" in the 80s and Kicked Jobs out of Apple
Today engineers have to put up a fight to do anything resembling craftsmanship.
That their own products depend on it because they developer their products in Mac. And that the professional people they pretend they cater to depend on Macs, and steadily move away.
Unrelenting bad press. People talking about nothing else but the decline of their software quality. We can already see that with the recent debacle which caused executive shuffling at the top of the company.
"Bad press" for their declining software quality is like people complaining there's no iPhone mini/SE anymore. Apple just doesn't give a fuck. They've joined the rest of the flock at chasing fads and quarterly bottom lines.
It was already the same story with AirPower (the wireless charging mat). They've pre-announced it, even tried to upsell it by advertising it on the AirPods packaging. It just turned out physics is ruthless.
TBH I've been increasingly sceptical about voice assistants in the "pre-AI" era. I sold my HomePods and unsubscribed from Apple Music because Siri couldn't even find things in my library.
I have almost the opposite problem this year. I tell the HomePod to turn the office lights on, it sometimes interprets this as a request to play music even though my library is actually empty, and the response is therefore to tell me that rather than turn on the lights.
Back in the pandemic, same problem with Alexa. Except it was in the kichen, so it said (the German equivalent of) "I can't find 'Kitchen' in your Spotify playlist" even though we didn't even have Spotify.
This is a solved problem since ~1970 -- they're just not spending enough time on it.
No idea what changed, but it sucks.
And that hardware needs to be coupled with solid software to hook and keep people on this computer. So they can take more time to create more compelling upgrades and sand off more edges.
I think they need to desync all their OS's and focus on providing better releases. There really is no benefit to spending the day updating your Mac, phone, tablet, appletv, and HomePod. Especially when there are no good reasons to update. I feel like Apple became far to addicting to habit and routine that it's become more important to keep that than deliver product. Apple Intelligence is a good example of that.
It's related to why companies with great marketing and fund raising but mediocre or off-the-shelf technology often win over companies with deeper and better tech that's really innovative. Innovation and polishing takes work that subtracts from the time available for fund raising and marketing.
Perhaps the real challenge isn't balancing innovation and marketing—it's creating a culture that genuinely rewards bold ideas and meaningful risk-taking.
Imho, this is the wrong takeaway from parent's point.
Bureaucracy rewards many things that are actual work and take time. (Networking, politicking, min/max'ing OKRs)
Creativity and innovation are rarely part of the list, because by definition they're less tangible and riskier.
A couple effective methods I've seen to fight the overall trend are (a) instill a culture where people succeed but processes fail (if a risky bet fails then the process goes under the spotlight, not the person) and (b) tie rewards to results that are less min/maxable (10x vs +5%).
> defaults favor desktop and server performance
Desktops are in S3 half the day consuming ~0 power. During use, electricity costs are so much lower than hardware costs that approximately nobody cares about or even measures the former. Servers have background tasks running at idle priority all day so the power consumption is effectively constant. Laptop and phone are the only platforms where the concept of "Linux power management" makes any sense.
"Idle" x86-64 SOHO servers still eat ~30W with carefully selected parts and when correctly tuned, >60W if you just put together random junk. "Cloud" works because of economies of scale. If there's a future where people own their stuff, minimising power draw is a key step.
If I already need a powerful machine for a desktop, why would I need a second one just so it can stay up 24/7 to run Miniflux or Syncthing? Less is more.
https://www.bee-link.com/products/beelink-ser9-ai-9-hx-370
I have the ser-8 model, and can confirm everything works under Linux. This one has an 80 TOPS AI thing, since you asked about llms.
Nahh, robustness comes from the time you can spend refining the product not from some magic property of a language. That can help but just a bit. There was no Swift in Snow Leopard. Nor there is not much Rust in Linux (often none) and even less (none) in one of the most stable OS available, FreeBSD.
They should just release a new version when the product is ready and not when the marketing says to release it.
I want Mac hardware but Linux software. The other makers build quality is horrendous. Especially in the 13inch segment which is my favorite. Using a pretty old laptop because there is no replacement right now.
The new Ryzen AI looks really interesting! Sadly there is no Framework shop for me to look at it and they not ship to Japan..
1. Display output from USB-C didn't work 2. Couldn't run Zotero 3. Couldn't compile Java bioinformatics tools 4. Container architecture mismatches led to catastrophic and hard-to-diagnose bugs
There were things that worked better, too (better task management apps, and working gamepad support come to mind). Overall, even though I only needed those things once or twice a week, the blockers added up and I erased my Asahi partition in the end.
I really appreciate the strides the Asahi project has made (no really, it's tremendous!), and while I would love to say that Linux lets me be most productive, features like Rosetta2 are really integrated that much better into MacOS so that I can't help but feel that Asahi is getting the worst of both worlds right now. I'll probably try again this summer and see what has developed.
Jokes aside, I had to wait years for Framework to finally allow shipping via a friend in Berlin. I think they ship to Sweden now—they seemed to have an unfortunate misunderstanding that they needed to produce a Swedish keyboard and translate their website before shipping here, which of course is poppycocks.
Out of curiosity, what are you basing this on? From having spoken to people who manage IT fleets, and being the person regular people ask for advice for what device to get, with the occasional exception (which Apple also had plenty of, cf. the butterfly keyboard), you get what you pay for. A 1k-1.5k+ Asus/Dell/HP/Lenovo will get you decent and good build quality.
The cheapest $500 Acer won't.
My work machine is an M2 Pro MBP and except the shitty input HW (compared to the golden era of Thinkpads/Latitudes without chiclet keyboards) and MacOS being quite bad compared to Linux, it completely trounces the neighbouring Dells that constantly need repairs (mostly the USB-C ports and wireless cards failing).
Got two "2k" Lenovos at 4 year intervals.
The first one worked fine but that model was known to have a weak hinge. Had to replace it three times.
The second one had a known problem that some units simply stop working with the internal display and the only solution is replacing the motherboard. My unit worked about a week for me. Seller refunded me instead of repairing because it was end of the line and they didn't have replacements.
Got a "2k" Asus ordered now, let's see how that goes :)
Compared to that, even the one emoji keyboard macbook pro that i had worked for years. The keyboard on those models is defective by design and kept degrading, and I still think Cook should take his dried frog pills more regularly, but the rest of the laptop is still working. Not to mention my other, older apple laptops that are still just fine(tm), just obsolete.
Where's a Thinkpad that can run Maya comfortably for a student? AFAIK they only have models with Quadros that have anything but student prices.
So I'm stuck with "gaming" models.
Besides my daughter likes the bling :) If only they could sell me something that doesn't die in a week...
And it still won't be on par with a $999 apple silicon air, or a MBP.
I've deployed latitudes, precisions, and thinkpads. They all still make tradeoffs that you don't have to deal with on the mac.
The X1 carbon is probably the "best" but, even with that - you are still getting a 1920x1200 screen unless you spend more than a MBP for the 2.8k display (which is still less than the 14" MBP, and costs more than an equivalent specced M4 pro). The trackpad is worse, the speakers are worse, battery life is worse, and they're loud under load.
They're all fine for a fleet where the end user isn't the purchaser, which is why they exist, but for an individual that doesn't want tradeoffs (outside of the tradeoff of having to use macOS), there's no other option on the market that comes remotely close to the mac. For someone that wants Apple silicon MBP level hardware but wants to run Linux, there are zero options.
The screen is the most egregious tradeoff though, the PC world is still adverse to HiDPI displays and even on high end models 1080p or 1200p is still the standard. I can excuse poor speakers, it is a laptop after all, if I really had to I can deal with fan noise, but I shouldn't have to spend more than a MBP to get a decent 120hz HiDPI screen with sufficient brightness and color accuracy.
I have a P1 Gen 7 and it’s fantastic. It feels premium, and it’s thin, light, powerful, has good connectivity and 4K OLED touch screen. I’d take it over Mac hardware any day.
The one snag I ran into was that when it was new, supporting the power modes properly needed a mainline kernel rather than the distro default. But in the grand scheme of things that's relatively trivial.
I have an M1 Macbook Pro from work and honestly I'm not tempted to get one for myself. I am tempted by the M3 and M4 beasts as AI machines, but as form factors go I'm just not sold.
But it is baffling how 1920x1080 (or 1200p) are still the "standard" elsewhere. If I want an X1 carbon, the best screen you can get at 14" right now is 2880x1800 (2.8k). Spec it with 32GB of RAM and it's clocking in at $2700, for a laptop that still has a worse trackpad, worse sound, and worse screen than a 14" MBP at $2399. And the Ultra7 in the thinkpad still doesn't beat the Mac, and it'll be loud with worse battery life.
There truly is nothing else out there with the same experience as an Apple Silicon MBP or Air.
So, my only options for the foreseeable future is wait for Asahi Linux, or suck it up and deal with macOS because at this rate I don't think there will ever be a laptop with the same quality (across all components) of the mac that can run Linux. The only one that came remotely close is the Surface Laptop 7 with the Snapdragon elite, but no Linux on that.
It would be kind of funny, but also very sad, if Apple guys mistook the copying of apple's worst behaviour - producing throwaway devices - as a sign of quality. Though I think we are there for years now with phones, I wouldn't expect such thinking here.
My point is that this system is not integrated in the way apple fans usually define the word. I'd claim it is not integrated at all. It is a regular PC (but with soldered ram), which is exactly like framework announced it.
There should be no need to sprinkle some apple marketing bs on that to make it attractive.
It’s absurd.
As someone who actually studied human computer interaction, and since I had to work with borderline unuseable macs multiple times in my career now, plus as someone seeing the utter failure of relatives in just using an iPhone (bought since "it is so much easier", now not even able to call from the car system since it is so buggy), the Apple popularity is absolutely a case where you have to look at external factors like social status. And if that translates to "the users are dummies" to you, then that's your interpretation. Plus yes, translating marketing/status concepts like a bogus "integrated" status absolutely is interesting, thus my intent to clarify whether that is really happening here (plus some criticism, admittedly).
Probably not worth it going further into this though, it will only derail.
That said, I think anything with kernel-level anti-cheat either does not run or runs poorly.
There is mention on the arch wiki about enabling multi-threaded compiles, but also I have read you perhaps dont even need to precompile them now and possibly get better performance as the JIT compiles via a different vulkan framework (VK_EXT_graphics_pipeline_library).
I disabled pre-caching (which effects the compile too afaict) and never noticed any stuttering, possibly past some level of compute it's inconsequential. I also noticed that sometimes the launcher would say "pre-compiling" but actually be downloading the cache which was a lot slower on my internet.
Certainly on my (very) old intel system with a GTX1060, Sekiro would try to recompile shaders every launch, pegging my system at 99% and running for an hour. I just started skipping it and never really felt it was an issue, Sekiro still ran fine.
I have a problem with Docker sockets while installing onto Bazzite, and didn’t care enough to look further into it.
First is the endless badgering to log in, LOG IN, LOGGGG INNNNN with an asinine Microsoft account. If you can tolerate that and actually get the OS running, you're wading through a wonderland of UI regressions and defects.
The default hiding and disabling of options is infuriating. Try showing content from your Windows computer on a TV, for example. You plug your HDMI cable in, and you can select the TV as an external monitor in a reasonably logical manner. Great.
But wait... the sound is still coming from the laptop speakers. So you go to Sound in the system settings. Click on the drop-down for available devices. NOPE; the only device is the laptop speakers.
So you start hunting through "advanced settings" or some such BS. And buried in there you find the TV, detected all along, but DISABLED BY DEFAULT. WHY??? Not auto-selecting it for output is one thing, but why is it DISABLED and HIDDEN?
This is the kind of shit I have to talk my parents through over the phone so they can watch their PBS subscription on their TV. The sheer stupidity of today's Windows UI isn't just annoying, but it's demoralizing to everyday people who blame THEMSELVES for not being "computer-savvy" or slow learners. NO; it's Microsoft's monumental design incompetence and user-hostile behavior.
Microsoft doesn't get the relentless excoriation it deserves for its miserable user experience. There's no excuse for it.
Meanwhile, my 7-year old laptop with Fedora on it I type this is wonderfully snappy and stable. I started to get tempted to actually switch back to a Mac just to get some predictability and stability, but I have avoided macs for years. (And - never having to deal with constant line ending issues)
All I hear from other co-workers is how their perfectly specced laptops lag with Windows. It's freaking Stockholm Syndrome here!
I am perfectly happy to use last-gen hardware from Ebay if it runs an OS that isn't begging me to pay for subscriptions and "developer fees" annually. My dignity as a human is well worth it.
* Battery life is a lie, especially since it drains almost as much battery closed as it does open.
...
Overall, I think I am probably going to switch back to a macbook after this, not being able to go a day without charging and your laptop always being on low battery is a bit anxiety inducing.
https://news.ycombinator.com/item?id=38206173
Metal isn't really on pair with Vulkan and DirectX in terms of relevance for graphics programming, the M chips aren't up to NVidia ecosystem, SYCL, the two major compute APIs for any kind of relevant GPGPU workloads, and thus don't really matter.
And gaming, well, even though all major engines support Metal, there is a reason DirectX porting kit is now a thing.
So why pay more for a lesser experience, and then there is the whole issue macOS doesn't have native support for containers, like Windows does (their own ones), and WSL is better integrated and easier to use than Virtualization Framework.
Proton is the acknowledgment of Valve's failure to entice game studios, already targeting Vulkan/OpenGL ES/OpenSL on Android NDK, Switch (which has OpenGL 4.6/Vulkan support), or on PlayStation (Orbis OS being a FreeBSD fork) to target GNU/Linux.
I rather have the real deal, not translations.
If you're playing the likes of Fromsoft/Resident Evil/Kojima games on a PC, be it Windows or Linux, you're not playing on the platform those games were designed for.
"Technical issues" has many meanings.
Unless you play benchmarks instead of games, and care about 8k/1200 fps of course.
There’s no such thing as “native”, all the things you’re talking about are translation layers for hardware instructions themselves, and the overhead for doing software based translation is significantly less than hardware accelerated virtual machines- and we as an industry love those.
The reason for this is because the translations are very cache friendly and happen in userland, so the performance impact is negligible, and the scheduler on Windows is so poor compared to Linux ones that it’s even common for games to perform better on Linux than on Windows.. Which is crazy when you consider the difference in quality of the GPU drivers.
I understand that you want it to “just work”, but that tends to be the experience anyway.
You can do what you want, it’s your life, but this is not a terribly good excuse. Valves “failure” is essentially rectified.
I will add though, that it’s actually Stadia that made linux gaming the most feasible, many game engines (all of the ones I worked on) were ported to Linux to run Stadia, those ports changed essential elements of the engine that would have been slow or difficult to translate; so when Proton came around quite a lot of heavy lifting had gone away. I only say this because Valve gets some credit for a lot of work our Engine programmers did to make Linux viable.
There is certainly such thing as native, one thing is the platform where the APIs were originally designed for and battled tested, and the other is other platform emulating / translating them, by reverse engineering their behaviours with various degrees of success.
Valve's luck is that so far Microsoft/XBox Gaming has decided to close an eye on Proton, and it will run out when Microsoft decides it has gone long enough.
Not sure, Unreal Engine is pretty popular though and Snowdrop is increasingly common for Ubisoft titles.
https://www.protondb.com/app/2842040 https://www.protondb.com/app/2840770/ https://www.protondb.com/app/365590
Unreal is almost worse because their first party tools (UGS, Horde) will not work on Linux, so you have to treat linux as a console, and honestly the market share isn't there to justify it.
Star Wars Outlaws
Natively Supports: Windows only
> https://www.protondb.com/app/2840770/
Avatar: Frontiers of Pandora
Natively Supports: Windows only
> https://www.protondb.com/app/365590
Tom Clancy’s The Division
Natively Supports: Windows only
----
You were saying?
Specifically, I worked on those games so I know what they natively support and how things transpired behind the scenes.
Proton has absolutely no hope of working without the changes we made because of stadia, the code we wrote was deeply hooked into Windows and we made more generic variants of many things.
The Division 1 PS4 release was significantly shimmed underneath compared to the win32 and xbox releases: this became much less true over time as porting the renderer to linux (specifically debian) made us genericise issues across the OS’s and when Div2 shipped we had a lot more in common across the releases; we didn’t rely on deep hooks into Microsoft APIs as much
Strange how you ported the renderer to Debian, and yet you couldn't even find a link to a game that has a native Linux support.
Was there ever a port?
> Proton has absolutely no hope of working without the changes
You keep saying this as the absolute truth, and yet at the time when Stadia launched Proton already had 5k working games under its belt.
Strange how Stadia is this monumental achievement without which Linux gaming wouldn't happen according to you.... and yet no one ever mentions Stadia ever contributing any code to any of the constituent parts of what makes Proton tick. Apart from the changes that engines supposedly made to work on a yet another game streaming platform.
There is a functioning version of The Division 1, Division 2, Avatar and Star Wars outlaws that run on Linux internally at Ubisoft.
Nobody will release it because it can’t be reasonably QA’d. (Stadia was also very hard to QA, but possible, as it was a stable target and development was essentially funded).
I’m not sure what your problem is; I said - as clearly as I can - that architectural changes to the engine were neccessary for proton.
I know this, for an absolute fact, because Proton was a topic when I worked on those games and it was not until Stadia (codename Yeti) was on the roadmap, and our rendering architect lost all his hair working on it - that Proton started to even function slightly.
I’m not shilling for Stadia - there’s nothing to shill for, it is dead.
Get over yourself, if you don’t like the truth then don’t start going in on me because my reality does not match your fantasy. Sometimes corporations do things accidentally that push other things forward unintentionally.
I just want to share my thanks to Stadia because I know for a concrete fucking fact that at least some of the AAA games I worked on would not function at all on Linux without Stadias commercial interference.
All I'm saying is that "it’s actually Stadia that made linux gaming the most feasible" statement is at best contentious because in reality gaming on Linux was already made (more) feasible when Stadia had only just launched.
And Stadia used the same tech without ever giving back to Proton at all (atl least nothing I can quickly discover). So the absolute vast majority of work on Proton was done by Valve which you dismissed as "when Proton came around" (it came around before Stadia) and "quite a lot of heavy lifting had gone away" (Valve did most of the heavy lifting).
That's the extent of my "problem".
> at least some of the AAA games I worked on would not function at all on Linux without Stadias commercial interference.
So, not "actually Stadia that made gaming feasible on Linux" but "because Stadia used all the same tech, and there were possible commercial incentives early on until Google completely dropped the ball, bigger studios also invested in compatibility with the tech stack"
Stadia did a lot to help by being a stable target and by being seen as commercially viable. Google also helped a lot to aid developers, not just financially.
That they didn’t contribute code to proton doesn’t factor at all, I just hate to see people not get their dues for their part in the prolification of Linux gaming- because I saw it first hand.
You are labouring under the delusion that I’ve implied Proton did nothing, no, they levied a lot of existing technology and put in a lot of polish. They were helped by Stadia, by Wine, by DXVK and others.
They didn’t do it alone, that doesn’t minimise their contribution, it contextualises them.
Also: Stadia ports of games were native, they did not use proton- it was architecture changes of the games themselves that made proton work better- not Google making proton itself function better.
That proton was running some games is a weird revisionist take, very few AAA games ran at all, those that did were super old and there was always some crazy weird bugs- proton got better but also AAA games coalesced into conforming to linux-y paradigms underneath better- so support got better much quicker than expected. You can even see this if you track the “gold” released games over years, some of the worst supported games for Proton are from 2015-16; before stadia but after game complexity started rocketing up with next game engines of the day.
Hope that helps, because honestly this conversation is like talking to a brick wall.
Oh, you very much minimised their contribution. From "when Proton came" (again, Proton came before Stadia) to "Stadia made gaming feasible on Linux" (when Proton made it feasible before Stadia)
> Also: Stadia ports of games were native, they did not use proton- it was architecture changes of the games themselves that made proton work better- not Google making proton itself function better.
So, Stadie games were Linux ports. But as a result of this there are still literally no Linux ports. None of the tech behind Stadia ever made it back into software behind Proton. And "native stadia ports" are somehow responsible for more games that target Windows and DirectX to run better via Proton
> That proton was running some games is a weird revisionist take
Funny to hear this coming from a revisionist. I literally provided you with links you carefully ignored
--- start quote ---
A look over the ProtonDB reports for June 2019, over 5.5K games reported to work with Steam Play
https://www.gamingonlinux.com/2019/07/a-look-over-the-proton...
--- end quote ---
> You can even see this if you track the “gold” released games over years, some of the worst supported games for Proton are from 2015-16; before stadia but after game complexity started rocketing up with next game engines of the day.
Or because the actual heavy lifting that Valve did with Proton paid off, and not the nebulous "native ports" and code that never saw the light of day.
> because honestly this conversation is like talking to a brick wall.
Indeed it is.
I play most of my games in a window and switch away a lot. A million years ago when I was still playing world of warcraft, the system overall was much more responsive on the same hardware with wow on wine on linux than with wow natively running on windows :)
> it’s actually Stadia that made linux gaming the most feasible
Stadia was the most predatory gaming offering aside from IAP games, sorry. Buy your games again on top of the subscription? Lose them when Google cancels the service? No thanks.
Nvidia's GeForce Now was a lot more honest. Pay for the GPU and streaming, access your owned games from Steam. I'm not using it any more so I don't know how honest they still are, but I did for like a year and it was fine(tm).
The fact that Stadia advanced wine compatibility is great, but technical reasons aren't the only reasons that make a service useful to your customers.
I worked closely with productions using proprietary game engines, I feel qualified in stating that Stadia had an outsized impact on our development process in a way that helped proton succeed.
That you don’t see it as an end user, is exactly my point.
You don't have to be a chef to judge what's coming out of the kitchen.
What is the objective impact of Stadia which at its height had a whopping 307 titles [1]? At the time of writing ProtonDB lists 6806 titles as "platinum, works perfectly out of the box" and 4839 games as "gold, works perfectly after tweaks". Steam Deck alone has almost 5x the number of games with "verified" status [2].
What games are being made for Linux thanks to Stadia, and don't just target DirectX and run through Proton? How many Stadia games were ported to Linux thanks to Stadia?
Also, to put things into perspective. Proton was launched in 2018. Stadia was launched in 2019.
In 2019 there were already over 5000 games that worked on Proton. [3]
In 2022 there already were more games with verified status for Steam Deck than there were games for Stadia, and 8 times more games verified to work by users [4]. Stadia shutdown was announced half a year after the article at [4].
Stadia had zero impact on gaming in general and on gaming on Linux in particular as judged by the results and objective reality. Even the games you showed as examples don't support Linux, only target Windows, and are only playable on Linux through Proton [5]
> I feel qualified in stating that Stadia had an outsized impact on our development process in a way that helped proton succeed.
> That you don’t see it as an end user, is exactly my point.
It's strange to claim things like "when Proton came along" when Proton was there before Stadia and already had over 5k games working in the year when Stadia only just launched.
It's strange to claim outsized impact on development process when there are no outcomes targeting anything even remotely close to Linux development, with studios targeting Windows as they have always done.
It's strange to claim Stadia had outsized impact when none of the work translated into any games outside Stadia. When Stadia did not contribute any significant work to the tech that is running Proton. In 2022 they even started work on their own emulation layer that went nowhere and AFAIK never contributed to anything [6]
It's strange to claim that "it's actually Stadia that made Linux gaming feasible" when there's literally no visible or measurable impact anywhere for any claim you make. Beyond "just trust me".
[1] According to https://www.mobygames.com/platform/stadia/ According to wikipedia, at the time of shutting down it had 280 games, https://en.wikipedia.org/wiki/List_of_Stadia_games
[2] https://www.protondb.com/dashboard
[3] https://www.gamingonlinux.com/2019/07/a-look-over-the-proton...
[4] https://www.protondb.com/news/how-many-games-work-on-linux-a...
[5] https://news.ycombinator.com/item?id=43503018
[6] https://www.gamingonlinux.com/2022/03/google-talk-about-thei...
You don’t know how the sausage is made just because you ate a hotdog.
Maybe you should consider things more carefully before making yourself look like an idiot on the internet and simultaneously raising my blood pressure.
Proton is a lesser implementation of Windows API, sure, but Windows itself is a lesser implementation of an operating system for power users.
The initial approach of runtimes did help but it's still has its limitation.
If now a studio just need to test their game under a runtime+proton the same way they would test a version of Windows to ensure it's working under Linux it's a win/win situation. Proton becomes the abstraction of the complex and diverse ecosystem of Linux which is both its strength and weakness.
Another solution would have been everybody using the exact same distribution which would have been way worse in my opinion.
And who knows, maybe one day Proton/Wine would be the Windows userland reference and Windows would just be an implementation of it :D
I thought only Apple had a distortion field.
Most of HN seems to think using a web browser as a translation layer is a good idea, yet they complain when games use a translation layer.
I miss the days of native apps with Internet protocols, and USENET discussions.
A web site makes for a crap application and the reverse.
Yes?
Anyway Linux is liberating, Fedora Desktop is great, no ads in the OS, a Software Store/Installer I actually like to use, curated by usefulness instead of scam Apps. All my Windows Steam Games I frequently use just worked, I have to login to X11 for 1 title (MK11), but everything else runs in the default Wayland desktop. Although I'll still check protondb.com before purchasing new games to make sure there'll be no issues. Thanks to Docker, JetBrains IDEs and most Daily Apps I use are cross-platform Desktop Web Apps (e.g. VS Code, Discord, Obsidian, etc) I was able to run everything I wanted to.
The command-line is also super charged in Linux starting with a GPU-accelerated Gnome terminal/ptyxis and Ghostty running Oh My Zsh that's enhanced with productivity tools like fzf, eza, bat, zoxide and starship. There's also awesome tools like lazydocker, lazygit, btop and neovim pushing the limits of what's possible in a terminal UI and distrobox which lets me easily run Ubuntu VMs to install experimental software without impacting my Fedora Desktop.
Image editors is the one area still lacking in Linux. On Windows I used Affinity Designer/Photo and Paint.NET for quick edits. On macOS I use Affinity & Pixelmator. On Linux we have to chose between Pinta (Paint.NET port), Krita and GIMP which are weaker and less intuitive alternatives. But with the new major release of GIMP 3 and having just discovered photopea.com things are starting to look up.
Xerox PARC is the future many of us want to be in, not PDP-11 clones.
Sure you can happily avoid the command-line with a Linux Desktop and GUI Apps, although as a developer I don't see how I could avoid using the terminal. Even on Windows I was using WSL a lot, it's just uncanny valley and slow compared to a real Linux terminal.
It is like praising Ratatui for what Turbo Vision, Clipper and curses were doing in 1990's, if I wanted that I would kept using Xenix and MS-DOS.
GUIs are very useful but they are not clearly better (or worse) than CLIs.
It's not a weird flex. Weird flex is this: "The command-line is also super charged in Linux starting with a GPU-accelerated Gnome terminal/ptyxis and Ghostty running Oh My Zsh" and then listing a bunch of obscure personal preference tools that follow trends du jour.
And you just alias them, so you can keep using the core utility names to use them.
As if Vulkan had relevance to graphics programming.
> and WSL is better integrated and easier to use than Virtualization Framework.
you don't need WSL on MacOS because, well, MacOS is already a *nix environment.
Right up until you need Linux syscalls. If you're doing anything with containers it's an annoyance.
It surely has on 80% of a mobile platform, and on a small handset from this little japanese games company.
> troupo 3 hours ago | parent | context | flag | on: Apple needs a Snow Sequoia
> Metal isn't really on pair with Vulkan and DirectX in terms of relevance for graphics programming
As if Vulkan had relevance to graphics programming.
> you don't need WSL on MacOS because, well, MacOS is already a *nix environment.
Agree if everything one wants out of it is classical UNIX experience, that breaks down when having to work with containers and kubernetes locally.
And which platform brings in more money?
> and on a small handset from this little japanese games company.
And not on PS, not on XBox, not on PC (that is, no first-party support).
But since I'm not gaming I cannot imagine going back to Windows. On the other hand I'm quite enjoying Linux...
> So why pay more for a lesser experience
...however, with few exceptions, I haven't used mouse in decade... and I haven't found anything like MBP's touchpad yet. Maybe I just need to do better research.
https://liliputing.com/argon40-is-making-a-raspberry-pi-cm5-...
Have a look at this sample code: https://developer.apple.com/documentation/virtualization/cre...
I wish Apple would just fix Spotlight. They don't seem to think it's worth fixing.
https://www.voidtools.com/
I could be wrong, but apparently Spotlight is the service that drives this kind of file system watching. I think macOS has a lower-level inotify-style file system event API, which should be unaffected, but Finder and these other apps apparently use Spotlight. I really wish I had a fix, because it's just crazy having to constantly "refresh" things.
You went through the effort to show some UI when something I am looking for may not be there because indexing is paused... but you didn't think to just unpause the indexing so that I can find it? I feel like I am being spit on, "Yeah, you not finding what you are looking for? I know, I'm not even trying"
https://www.alfredapp.com/
[1] https://qsapp.com
Probably not.
Who the hell thought integrating internet search is a good idea - because "aösldkfjalsdkfjalsdkfj" just as everything else is a valid search result in Spotlight now showing me "Search for aölsdkfjöalsdfjasdlfkj in Firefox"...
I still have some 10.6.8 install media for both server and client. Truly loved them both.
Another highlight of that job was selling a green iPod Nano to "John Locke" from LOST
The most ridiculous thing that happend to me was in the early days of the Apple Store in SOHO I stopped in to see if I could just buy RAM.
The music was loud so it was like I was speaking loudly to be heard and asked for RAM and the they thought I was asking if I could buy a gram.
Loved Snow Leopard too, & was shocked by how bad Lion was in comparison. Glad they got back on track after that.
Wasn’t 7.5.3 the worst of the string of terrible releases between 7.5 and 7.6? In my memory 7.5.5 was much better, but I still preferred 8.1.
You can only cycle windows in one direction even if you try to do some remapping
Choosing keyboard languages hides a lot of options. Once you understand you need to click on English US to see more detailed options then you get them all, UK, Canadian... Then it's unclear which keyboard layout is currently selected and how to select one from the list you made.
I can't fathom how a DE whose is all about human machine interface guidelines whatever and supposed to be the epitome of UX can't figure out basic stuff about discoverability and clarity
Keyboard layouts are a pain, but there are some solid extensions that clean the flow up and may be upstreamed into GNOME at some point.
It's all opinions, but boy, compared to the mess that is macOS and iOS regarding discoverability ... I'll take GNOME any. day.
The volume and power icons on the top right is actually one button and hides other option like screen lightning volume and wifi etc. If at list they had made a three vertical dots/stacked bars and is the convention for hamburger menus...
From what I heard GNOME devs do not like change and it sucks to be a GNOME extension developer, a quick google search seems to confirm that so it casts some doubt about them up-streaming any of them but maybe you know better. Has it ever happened to other extensions ?
https://discourse.gnome.org/t/developing-gnome-shell-extensi... https://www.reddit.com/r/gnome/comments/pvvku5/why_do_extens...
Haven't really used MacOS or iOS more that five minutes so I can only trust you on that.
On the other hand for example, it is very easy to remap CapsLock to Escape on MacOs. Just go to Setting --> Keyboard and you easily find the option. GNOME ? No, not in settings. Wait I have to use an app called gnome-tweak ? Ok it's in "Advanced keyboard otions" --> Opens a big list of loosely classified options. Oh well it was in miscellaneous category.
I don't know if the CapsLock -> Escape switch is on a roadmap somewhere, but that is a little bananas. That said, my partner comfortably uses GNOME every day to browse the web and manage some files. Has she EVER wondered how to remap CapsLock? No. The people who do want to? Google can give you the answer pretty quickly. Not saying it's good UX, but GNOME balances a lot of use cases, and as this thread suggests, I think they've actually (with a LOT of complaining from engineers and power users) kept that balance pretty damn well to the point where I haven't been surprised by GNOME is a long time, and seems to slowly and progressively get better.
And yes, whoever jumps in here with their own papercut story, I know there is pain in not being the primary audience for software. But honestly, at least I'm in the same Venn diagram with my partner. The primary audience for macOS or iOS now appears to be ... I don't even know anymore. Used to be content creators, now it seems like even Apple doesn't actually know who uses their computers.
Feelings shared, if only Gnome would provide this column-based file navigation that I miss so much
The things that most frustrate me about Macs is that they've violated the never spoken but always expected "it just works" in so many ways. Things like how Thunderbolt Displays containing a USB hub which are Apple-certified handle re-connection to a Macbook, should "just work", but require fiddling every time. That's just one of numerous examples I could come up with.
Apple historically was probably the best company in the world in understanding the full depth of what "User Experience" means, and it seems like they've really retreated from this position and are regressing to the mean.
Here are some data points I collected at the time:
https://blog.rongarret.info/2009/08/snow-leopard-is-disaster...
https://blog.rongarret.info/2009/09/esata-on-snow-leopard.ht...
In retrospect Snow Leopard deserves the love it eventually got, but at the time it was not entirely clear.
> Apple's software quality (either in terms of polish or just plain QA) has steadily decreased.
Amen to that.
OMG this one drives me bonkers. If anyone out there knows how to turn off internet results, please share!
You’re welcome :-)
I struggle to imagine the software design that works so poorly.
Then you do a search for .jpg, and get NOTHING. But only sometimes. Other times it'll work.
And it makes no sense whatsoever. If "foo" matches "foobar", so should "foob". I honestly don't know how the hell can they still f up such a simple piece of technology in 2025.
https://www.voidtools.com/en-uk/support/everything/
So... when the hits include six identically-named files, you can't eliminate ones that you know are wrong (on a backup volume or whatever). The level of stupidity here is just mind-boggling.
I guess you do know the path is shown at the bottom of the window if you select the filename in the list of results?
Absurd.
To select, just press on the item.
To hover, press and hold for at least 2 seconds.
To get a list of options, press and hold for at least 2.5 seconds, but not more than 3.5 seconds.
To delete, press and hold for 3.6 seconds, but not longer than 3.9 seconds.
To save, press and hold for 4.1 seconds. Pressing and holding for exactly 4.0 seconds activates the archive action. Pressing and holding for 4.2 or more seconds sends the item to the blocked list.
To retrieve the list of items in the blocked list, press and hold and simultaneously press the volume up and volume down key.
To delete all items in the block list, press and hold and simultaneously press the volume up key only.
To completely reset your device, press and hold and simultaneously press the volume down key only, whilst holding the device in a completely vertical plane, and rotating clock-wise and counter-clockwise, smoothly, at precise 2.35619 radians every 30 seconds.
To trigger the emergency call feature, drop the device at an acceleration of no less than 9.6m/s and no more than 9.7m/s
/s (kind of)
The whole point is that secret hotkeys are design dereliction.
That’s a bit of a problem when discussing problems of normal users with power users, because they don’t even realise how what they’re doing is actually not what normies do.
I’m inclined to agree that hotkeys in MacOS are hard to discover, but cluttering the interface with stuff many users simply do not need cannot be the correct answer.
It also doesn't allow you to sort results by location, as you could if it were a column.
I also just tried it in Spotlight and Finder, and it did nothing. Which I consider a relief, because undiscoverable bullshit is worse than the feature not existing.
Special mention to all text input fields in macOS having Emacs-style shortcuts.
https://support.apple.com/en-gb/guide/mac-help/mchlp1008/mac
I agree that discoverability could be better, but macOS has pretty consistently had hidden power user shortcuts and modifiers, to keep the basic workflow streamlined/simple for those who don't need it.
And I don't buy the "keeping things simple" excuse for secret hotkeys in other areas. Falling back on gimmicks like undisplayed hotkeys and "long presses" and "gestures" is lazy abandonment of the design task.
I hate this "saving the user from complexity" lie. It's hypocritical: The "non-power" user isn't going to go looking for these options in the first place.
Finder search is a great example. A "non-power" user isn't going to right-click on the column headings in the results and try to add "path" as a column. So how does it help that user to deny everyone else the ability to add it?
Apple mocked IBM for needing a thick user manual back in the day. To suggest that anyone (especially anyone on this site) should have to read documentation to use perform a basic file search (in a GUI, no less) is apologism to the extreme.
At least there's quicksilver
However I'm done with Apple. I think it's a decision - not "reasoning". That decision takes time and is painful. It's also a decision specifically against "the best" ecosystem available in favor of something "ok".
Not only they repeatedly disappointed my expectations - they just suck as a company (in my opinion). It's not about being less innovative for decreasing software quality, they have done so much for the market, that I think GNOME wouldn't even exist as it is without them... Its about sealing off every inch of their software and hardware they can. No repair without paying... Making RAM and SSD upgrades ridiculously expensive, you cannot even put default NVMe drives into a mac mini - everything is proprietary. Even their sensors have serial numbers to prevent hibernating if you change them out without "hacking" the firmware.
Hardware-wise I have high hopes for framework working with AMD - although they did not address the issues I'd suggest (speakers, lpcamm2), they're constantly improving without breaking their promises. This is hopefully not going to change when they get bigger.
OS-wise I'll stay on Linux. After a long journey going from Ubuntu to Debian to Fedora using GNOME, KDE and even NixOS with Hyprland for a short period, I gained enough knowledge required to really enjoy Linux. System76 is working on COSMIC, which could be pretty amazing, once it is released.
In case anyone would like to try my current Linux config, I'm constantly working on an "install everything" script (pretty early stage):
https://github.com/sandreas/zarch
HF ;)
You might not want one though.
You don't get nearly as much compute as you would with 6 GPUs, but it also uses less power than a single GPU.
This issue with spotlight is so bad. I use the switcher to pull up my Downloads or Documents directories and half the time it can’t even find them!
And so many tiny thumbnails wedged into the too-narrow System Settings window.
FWIW you can massively improve things by just disabling the internet results. It's easily done in the System Preferences
I think the decline of software went hand-in-hand with the decline of the native indie Mac app. They still exist, but when I started with the Mac (2007), there was a very rich ecosystem of native Mac apps. Most stood head and shoulders above their Linux and Windows counterparts.
Apple has nearly destroyed that ecosystem with: race-to-the-bottom pricing incited by the App Store; general neglect of the Mac platform (especially between ~2016 and Apple Silicon); and a messy reactionary toolkit story with Catalyst, SwiftUI, etc. The new toolkits seem to imply that Apple says that it's the end of AppKit, but most SwiftUI applications are noticeably worse.
With their messy toolkit story and general neglect, developers have started using Electron more and more. Sure, part of the popularity is cost savings, since Electron apps can be used on multiple platforms. But part of it is also that a Catalyst or SwiftUI app is not going to provide much more over an Electron app. They will also feel weirdly out of place and you become dependent on Apple working out quirks in SwiftUI. E.g. 1Password tried SwiftUI for their Mac app, but decided in the end that it was an uphill battle and switched to Electron on Mac instead.
I recently bought a ThinkPad to use besides my MacBook. Switching is much easier than 10 or 15 years ago, since 80% of the apps that I use most frequently (Slack, Obsidian, 1Password, etc.) are Electron anyway. Even fingerprint unlocking works in 1Password. I was vehemently anti-electron and still don't like it a lot, but I am happy that it makes moving to a non-Apple platform much easier.
Now, ios gets the executive attention and it will generally get the best developers assigned to it, and the Mac has to live with the scraps.
1Password had a beautiful native Mac app that works to this day. Even assuming SwiftUI is actually bad, why did they have to migrate at all? What was wrong with the existing app?
I'm not disagreeing with the opinions on Apple software quality, but I think the 1Password case is more down to their taking of VC money and having to give (JS) devs some busywork to rebuild something that worked perfectly well.
It didn't work on Windows and Linux desktops.
While it's true for 1Password, there are other password managers. KeePass is great for local password database files if that's what you're after.
I'm on the same boat here. Something is driving me away from my MacBook M1(Pro? Don't even know). I have a gut feeling that it's macOS but can't really put a finger on it yet.
Bought a heavily used ThinkPad T480s (from 2018) and replaced almost every replaceable part of it, including the screen. Being able to replace many parts easily is a nice touch since I am using MacBooks since 2007 exclusively. Guess that's why I somehow overdid it here. Slammed Pop!_OS 22.04 on it and I'm very pleased with the result. The first Linux desktop I actually enjoy since trying SuSE 5-something. Pain points are teams (running in browser), bad audio quality with AirPods when using the microphone and cpu speed and heat. I guess one has to stop using Apple silicon in laptops to realize how amazing these processors are.
Intel CPUs from that era were quite bad and everyone has upped their ante since then. I was thinking about getting a second hand from ~2021-2022, but my wife convinced me to get a new one, so I got a Gen 5 T14 AMD. It has a Ryzen 7 Pro 8840U and I rarely hear the fans, mostly only when Nix has to rebuild some large packages (running NixOS unstable-small).
Similar for me but started in system 7.
It’s just lucky Windows has got worse faster.
Similar for me but started in system 7.
It’s lucky for Apple that Windows has got worse faster.
I'm stuck on a MBP because it's the only laptop with a great screen, speakers, and battery life. Meanwhile my keyboard keys keep getting stuck after a year of usage, and OSX is garbage. Soon as there is similar hardware I can load Linux on, I'll be insta-switching.
I would love to finally get out of Apple ecosystem, I just don't have any decent alternatives right now. Hopefully next year.
IIRC some competitors are starting to offer a few laptops with ARM processors, I think Samsung has a few. How do you feel about those?
Of course today it would be insecure, missing security patches etc. SSL...
I bought an M1 Max that is now almost 4 years old and it still feels new to me. I can't really imagine a change that would happen in the next 2 years that would make this thing feel slow where an M3 would feel sufficient, so I'm curious to see if Apple really does just go hardcore on forced obsolescence going forward. I have a few M series devies now, from M1 to M3, and I honestly cannot tell the difference other than export times for video.
I can imagine some kind of architecture change that might come with an M6 or something that would force an upgrade path, but I can't see any reason other than just forcing upgrades to drop support between M1-M5. Maybe if there is a really hard push next year into 8K video? Never even tried to edit 8K, so I don't know. I'm guessing an M1 might feel sluggish?
Yes, they can theoretically perform better, but only when plugged into mains power, and creating so much heat and fan noise that the experience really isn't good.
Don't think there's anything out there that will outperform the GPU of an M-series Mac without consuming way more power and producing problematic levels of heat+noise.
In fairness, Apple to do tend to continue to release critical security patches for older versions.
I suspect that it will be AI features that push Apple into deprecating older hardware. But I also hope that the M series hardware will be supported a bit longer than the intel hardware was. Time will tell.
How are the keycaps doing? Mine looked awful after about 2 years of relatively light use, developing really obvious ugly shiny patches (particularly bad on the space bar), quite a letdown on an otherwise great machine.
(Realised that you can actually buy replacements and swap them yourself, via the self-service repair store, so have replaced them once, but am starting to notice shiny patches again on the new set)
Using an x86 laptop in 2025 is like using a flip phone.
From an OS / software perspective:
Have a "core" macOS that has none of the apps / integrations are baked in at an OS level.
You install the things you want, how you want - eg iMessage, Mail, and then iCloud if you want to sync it, and Photos etc.
Have a slim, fast, stable OS that I can just turn on and get going with.
From the hardware perspective, I made this comment a little while ago but what I want to be able to choose is:
- Device: Watch, iPhone, iPad, MacBook, iMac, Mac
- Size: Mini, "Normal / Default" (Air), Max
- Processing Power: "Normal / Default", Pro, Ultra
- And maybe storage.
That way I can go and buy a MacBook Pro (13"?), or a MacBook Max Pro (15"), or a MacBook Mini (11"), or a normal iPad Mini Ultra, or an iMac Mini (21"?), or a Watch Pro, or a Mac Max Ultra etc.
Device + Size + Power.
It's kinda there, but not quite.
For example, can you remove Chess from MacOS? Nope! Why? What I found on Reddit, it seems because it's integral part of MacOS somehow and I am a bad person for even asking, somehow.
Each part alone might not be large but together it starts to become an annoyance.
Also bloat is not just about disk space but also cognitive load and clean interface.
Those are not bloat, those are core features of a computer for 99% of users who are not developers.
There already exists a platform which is unusable for normal people and great for developers, it's called Linux.
There already exists a platform which is great for corporate and hell for normal people, it's called Windows.
So why aren't we allowed to keep the only computing platform which is good for normal users?
There's no need for a separated core version - just give back control to the user. But honestly, I don't know what would need to happen so we could get it - it feels like it's a lost cause against corporations. There's of course Apple-EU situation where you can remove applications, set defaults, install additional app stores but this is still limited to that market and happen way too late and too slow.
Hard agree with this. I sometimes have to boot up a windows laptop to play Minecraft with the kiddo, and it never stops reminding me how little I know about Windows now, how counter-intuitive everything is, how everything feels designed for a user whose mind I cannot comprehend.
How many "control panels"? How many places are there to adjust audio device properties?
(... speaking as another dad just trying to play with my kid.)
I figured that I make a six-figure salary as a software developer, I can afford $2/month so that I don't have to fucking become a sysadmin for a game server my child depends on.
There are two editions, Java and Bedrock. Java is the original, available on PC and Mac, and supports programming-like technical play and mods. Bedrock is Microsoft’s reimplementation, available on all devices except Mac, and supports emotes and microtransactions. Other than that they’re largely the same game, and buying either gives you both versions. Realms supports both, but a server is one or the other, not both. There are also other managed hosting providers for Minecraft (both versions), but Realms is probably easier and cheaper for you. Java version has performance problems, but mostly because Microsoft’s code is inefficient, there are a few mods (also written in Java) that everybody uses to fix performance without affecting gameplay.
You can access most settings by Windows + "yourquery".
It is indicative of a failure, not a solution in and of itself.
FWIW, search as a UI isn't a bad thing, Cmd + Space is the main way I launch apps on macOS (or Win + "type whatever").
To be fair, it's hard to say whether the Settings app is more broken in Windows or macOS these days. I think I'd have to give the crown to macOS here on account of search itself being more broken.
For example:
I prefer keeping my hands on the keyboard, and typing cmd+space followed by mouse is so much faster than finding the right pixels to click through in menu trees when I want to adjust my mouse sensitivity.
The UI fail is if search is required to find the setting every time you need it, because categorization and/or navigation is broken otherwise.
As to keeping your hands on the keyboard, that's an argument for having proper keyboard support in any UI, complete with hotkeys and shortcuts. The big difference between these and search is that the former is (if properly done), consistent and predictable. So e.g. when the app adds new things in the future, your existing key sequences will still do the same thing they did before.
To take your specific examples, if I do Cmd+Space, "mouse", Enter on my system, it will bring up LinearMouse, not system mouse settings.
On my iphone, I have one page of apps, everything else in the app drawer, and use the search all the time. It often gets what I want in one or two chars.
It also has the benefit of being roughly bilingual (English + Installed language) and being there even in machines not setup for you
I can get my mom's computer, fully set in spanish, and I can win + "query", into settings, programs and tools to setup whatever she needs
I admit I honestly have no idea where the system settings are located as I haven't pressed the start button in ages, but the same applies to MacOS as I would use spotlight there as well.
The search doesn't even work all the time. Sometimes it won't do fuzzy search, sometimes typing "bluetooth settings" will do a Bing search, some other time it will open a PDF, and so on.
And for reasons I don't understand, why is the window itself not resizable?
It would be fine if the settings available were actually useful or at least could bring me to some tool that does it better. I get no meaningful report of what's eating my batter and why every time I open my MacBook it's dead. And if I want to change the actual resolution of my display I'm given just a list of scaling options pretending to be resolutions. Oh, want to set a specific resolution or refresh rate? You have to do some stupid kinger king foo of option control something _before_ you click on this dialog. I get the criticism about the Windows settings app and legacy power tools (I think this has largely been solved anyway), at least they exist and allow me some iota of control over my computer
It blows my mind that when right-clicking on a file in file explorer, the 'delete' option is hidden in a sub-menu under 'more options'.
I agree that having "more options" to begin with was a jarring experience coming from windows 10 though.
Out of old habit I always use shift + DEL key and did not notice it's in the top row now.
It also makes way more sense.
Probably helps that I installed the IoT LTSC version, but still, apart from the task bar being stupidly in the middle (thankfully there's an option to move it to the left), I've had zero issues.
I even added a network printer and it found it quickly, and added it quickly and successfully, which is a feat I don't think I've seen happen on any OS ever.
The context menu is a clear improvement on the old one (which you can still get to with one click).
1) https://github.com/Raphire/Win11Debloat
It's the rule lest someone think you made a bad decision and you're regretting it. Even though it's an OS targeted for your grandmother, you must not let them see weakness.
At this point it's a joke. Either critique Apple or admit you can't without also bringing up some other OS. It's weird.
TL;DR What people remember fondly is not Mac OS X 10.6.0, which was in fact very buggy, and buggier than 10.5.8, but rather later versions of Snow Leopard after almost 2 full years of bug fixes.
See also "Snow Leopard bug deletes all user data": https://www.reuters.com/article/lifestyle/snow-leopard-bug-d...
The yearly release cycle is the problem. Apple needs "another Snow Leopard" only in the sense that I mentioned above, "almost 2 full years of bug fixes", although at this point, Apple has more than 2 years of technical debt.
My recommendation for people who don't absolutely need the latest features: Upgrade to the previous version of macOS when the new version is released. Sequoia is incredibly reliable 7 (soon to be 8) updates in.
I disagree with that part. ;-)
We wouldn't even be having this discussion right now if today's updates were incredibly reliable.
This is what’s being asked for in the article.
When Sequoia eliminated the ability to override Gatekeeper by control-clicking, it became clear to me that Apple is now employing a frog boiling strategy towards their ultimate goal -- more control of the software you can run on their hardware.
Trying to get the program to work with our Mac users has become harder and harder. These are all internal developers.
Enabling developer mode and allowing Terminal execution isn't enough. Disabling the quarantine bit works - sometimes - but now we're getting automated nastygrams from corporate IT threatening to kick the laptops off the network. I'm exhausted. The emergency workaround, which I tell nobody about, is way less secure than if they just let us run our own software on our own computer.
Developers pay exorbitant amount of money for much lesser value, and the idea of putting your teammates at risk to stick it to apple is kind of sad bordering with negligence from a business POV.
The cost is far far higher than the price.
I develop and distribute few free apps for macOS, and building / notarising is never a problem.
In most cases, just involving account management makes the corporate case 10x more of a PITA. Doing things in a corporate environment is a different game altogether.
Yes, you need to put keys on the build server for the "Developer ID Application" (which is what you need to distribute apps outside of AppStore) signature to work.
You do not need to give any special access to anything else beyond that.
Anyway, it is indeed more difficult than cross-build for Darwin from linux and call it a day.
It is ugly: https://hearsum.ca/posts/history-of-code-signing-at-mozilla/
I do this professionally, I maintain macOS CI workers for my employer. Apple doesn't make it easy.
For your personal needs, you do not need to pay anything for building and using apps locally.
It's about setting a higher floor for malicious actors than "random botnet residential IP + a captcha solving service". It's about proving some semblance of identity through a card number and a transaction that goes through without a chargeback.
As the case upthread shows, there's plenty to dislike about a system that inhibits running code built for personal use. And it's obviously neither foolproof nor without collateral damage. Reasonable people can debate if it's worth it. But it still ought be acknowledged that the motivations are closer to the reason you have to identify yourself and pay a nominal fee to drive a vehicle on public roads.
Since this isn't true, no acknowledgement required, it doesn't need to be a "major" profit center to magically become a benevolent feature
In particular, the security boundaries are nonsensical. The whole model of "notarization" is that the developer of some software has convinced Apple that the software as a whole (not a specific running instance) is worthy of doing a specific thing to the system as a whole.
But this is almost useless. Should Facebook be allowed to do various things that can violate privacy and steal data? What if the app has a valid reason to sometimes do those things?
Or, more egregiously, consider something like VSCode. I run it, and the fancy Apple sandbox helpfully asks me if I want to grant access to "Documents." The answer is really "no! -- I want to grant access to the specific folders that I want this workspace to access", but MacOS isn't even close to being able to understand that. So instead, one needs to grant permission, at which point, the user is completely pwned, as VSCode is wildly insecure.
So no, I really don't believe that MacOS's security model makes its users meaningfully more secure. At best, the code signing scheme has some value for attribution after an attack occurs, but most attacks seem to involve stolen credentials, and I bet a bunch just hijack validly-notarized-but-insecure software a la the VSCode example.
Notarization does some minimal checks, but is mostly about attaching a real identity so that maliciousness has at least some real-world consequences. The most obvious being that you lose the ability to get more apps notarized.
Last time I tried setting up an Apple developer license inside a large corporation, one that they paid for and not tied to me or my credit card, it was also a nightmare.
And yes, it's also on principle.
Permission that can be revoked for any reason, including being compelled by someone with more power than Apple.
Once signed, binary will work forever, you only need active subscription when you need to re-sign / re-notarise.
Do you have any evidence that it happened in any different circumstances at least once?
https://developer.apple.com/documentation/security/code-sign...
https://developer.apple.com/documentation/security/notarizin...
There are dedicated sections of the developer web forums:
https://developer.apple.com/forums/topics/code-signing-topic
https://developer.apple.com/forums/topics/code-signing-topic...
...and there's an apple developer support person, Quinn, who appears to be heavily if not solely dedicated to helping developers do binary signing/notarization/stapling correctly.
They have written a slew of Tech Notes about signing and notarization. Main TN is at https://developer.apple.com/documentation/technotes/tn3125-i...
Quinn also has their email address in their sig so people can just reach out via email without even needing an Apple account, or if they prefer more confidentiality.
I mean, come on.
Maybe because I'm using Electron framework which makes things more complicated, but I don't really understand why there's is a difference between different types of certificates (Developer ID, Apple distribution, macOS distribution) and I had to guess which one to use everytime I set it up.
Also why is notorization a completely different process from code signing, and requires completely different set of credentials from it. Seems odd to me.
Because they do completely different things. Signing is a proof that you were the one to write and package that software; notarisation is an online security check for malware. If I recall, you still sign but do not notarise when distributing to the Mac App Store.
Or maybe simpler, why can't Apple just do code sign and notarization with one single cli call, with one set of credentials?
Google Play does this under the hook, I don't even think about it. iOS is similar, Transponder app does everything in one go.
Edit: I haven't tested it yet, but it does seem that you can sign an executable with your own certificate (self-signed or internal CA-issued) however you can't notarize it. Right now, notarization is only required for certain kinds of Apple-issued developer certificates, but that may change in the future.
btw, for those who don’t want to search, Quinn’s signature states:
“ Quinn “The Eskimo!” @ Developer Technical Support @ Apple let myEmail = "eskimo" + "1" + "@" + "apple.com"
I once really urgently needed `nmap` to do some production debugging ASAP. Unfortunately, the security tools would flag this immediately on my machine, as I knew this from previous experiments. Solution - compile my own binary from sources, then quickly rename it. I assume that this "workaround" was totally fine for sec department. At least production got fixed and money kept flowing.
You were denied the tools to get your job done. You've put yourself at risk by applying an unapproved workaround.
Never ever do this (unless you hold substantial shares). Let the company's bottom line take the fall. If that's the only thing they care about, that's your only way to make the problem visible.
But the underlying SRE culture here is that, if you know what you are doing and have a functioning brain of a responsible person, you'd be forgiven a jump over the fence, if it means putting out a fire on the other side of it. We aren't kids.
I don’t get this at all.
I’d much prefer a team of highly empowered and highly responsible engineers than impotent engineers who need hand holding in case they make a mistake.
Engineers _should_ have leeway in how they resolve issues. As I read, though, you have a company policy which explicitly disallows the action you needed to take to fix the problem (if I misread, my apologies). Getting the stakeholders involved is the responsible thing to do when policies need to be broken.
Ideally, the way this kind of situation gets handled should be documented as part of a break-glass policy, so there’s no ambiguity. If that’s not the case, though, the business should get to decide, alongside the policy maker (e.g.: security), whether that policy should be broken as part of an emergency fix, and how to remediate the policy drift after the crisis.
If you’re all tight enough that you’re allowed to make these kinds of decisions in the heat of the moment, that’s great, but it should be agreed upon, and documented, beforehand.
By the way, I'm still burnt out. This work is stressful. Don't let it take away what's already scarce for you.
For binary patching: codesign --force --deep -s - <file> (no developer ID required, "ad-hoc signing" is just updating a few hashes here and there). Note that you should otherwise not use codesign as it is the job of the linker to do it.
Last time we did this I had to spend a week explaining to management that Macs could actually run software other than PowerPoint and it was necessary for our job.
The local workaround that we use is to just spin up a Linux VM and program devices from there. The less legal workaround is using WebUSB and I'm afraid to even tell the necessary people how I did it, because it's sitting out on a public-facing server.
Predefined value on current macOS's Gatekeeper is "move to Bin" instead of OK. Other option is Done - which cancels opening action. If you want to bypass that, you need to go to system settings > privacy & security and manually allow particular app there.
Who know what later updated will bring.
To run a non-motorized app requires you to open a separate app, navigate to the security section and select that you want to authorize the app to run.
Apple does not have any desire to make distribution of non-notarized binaries commercially viable.
And we've seen this change across all browsers. There no longer is a "continue" prompt for TLS issues. The result is, way fewer maintained sites go months with an expired certificate.
But here's my (unpopular) take as a GNOME user and using Fedora immutable distros + flatpaks -- I suspect Linux is going to go in a broadly similar direction. Maybe not soon (even flatpaks aren't universally acclaimed), but sometime.
I don't even mind that they've introduced a level on the totem pole that's above root. But on my computer, -I- should be the one at that level, not Apple.
the issue seems to be that you still believe this?
However, less corporate distros that mostly just ship built upstream software as-is since they don't have to support it for long periods (think Arch, Fedora, Void, etc) don't have that problem, so I expect we'll continue seeing them use traditional packages.
Ubuntu does the exact same thing with their snap repository, the Firefox apt package from Ubuntu is fake. At least Flatpak is a community-led project unlike snap.
You can limit the file system permissions of the app, like giving only access to downloads, so that if/when there’s a sandbox leak you’re fine. You can also disable various things, like webcam or mic, this way.
In addition, you can get perpetual updates to the latest version of your browser even on old, stable distros like Debian.
Linux is pretty diverse, there are still distributions out there that haven't adopted systemd.
It’s like criticism of the quality of Google search dropping. It has absolutely tanked, but it’s not because the algorithm is worse, it’s because the internet has grown orders of magnitude and most of it uses the same hyper aggressive SEO optimisation, such that the signal to noise ratio is far worse than ever before.
"Those who refuse to give up essential Liberty to purchase temporary Safety deserve to have to deal with the GNOME desktop user experience."
I miss macOS sometimes.
Kagi lets me completely block specific domains. If Google cared about quality they’d let you do the same.
We know this because the emails came out in discovery for one of the antitrust suits.
Suddenly the users file hierarchy started wherever the Home folder was located and it became an island of user controlled environment surrounded by complexity of computer operating systems.
The result I found overall well thought out but when the desktop became just a folder I felt the Mac moved from it’s simplicity embracing the complexity that was offered by windows.
It's amazing the rose tinted glasses people have about the original Macintosh environment. It was insanely janky and (unless you were ruthlessly conservative) insanely unstable by today's standards. By version 10.5 (Leopard) the modern UNIX-based MacOS was unequivocally superior to Classic MacOS in every metric other than nostalgia.
I also believe that the simplicity could have security as performant. The real advantage of the Unix layer is compatibility that the Macintosh was missing.
Are you trying to say that it’s possible for a system to be both simple and secure? Absolutely that’s the case, but with a trade-off — either it needs to restrict the user’s freedom, or be fully disconnected from the outside world.
The threats in the world are real and the internet doesn't help. I 100% agree that a network connection needs to be kept at a distance to make things simpler.
I think the power of language used to describe a system is where simplicity begins.
What I'm working on is creating a crisp line of delineation between "local" and "public" networks.
If by default after is on the "local" network auto-discovery is secure. If things are explicitly needed a user can publish them through physical manipulation to publish to the outside world.
The outside world can now be described using classic Users and Groups which is cultural easy to understand.
I'm trying to create an environment that focuses on making those 2 things plus a third element simple to understand and physically manipulatable.
The freedom I'm looking for is available on the "local" network. The "public" network is where our data is interchanged with the outside world based on our publishing. I don't expect people to interact with this layer much. I expect people to configure it for whatever institution/organization/government.
Most of the complexity I see in computing these days is market drive demand for eyeballs/clicks/...
I can't believe I even have to say this out loud. Look up enshittification.
Actively depleting the good-will they accumulated over the years definitely makes it worse. It's that harder to give the benefit of the doubt to a company also showing the middle finger to their Devs.
> Google
Giving priority to AdSense sites, fucking around with content lengths (famously penalising short stay sites), killing advanced search options. That's just thinking about it for 10s, but to me most of it is totally of Google's making.
sudo spctl —-master-disable
Are most people better off with Apple defaults?
And it’s not because the problem is “difficult”. It’s because for 20 years it has been claimed that this will be the “year of Linux on the Desktop” and it’s never been good enough for most people.
The second part of your post is incoherent to me, I can't tell what you're trying to say.
The problem with Linux is that, while it’s very good, it’s different.
Nobody actually cares how intuitive something is, at least not in absolute. People will still say Windows is intuitive. Pretty much nothing in Windows, from the registry to COM to IIS to setting/control panel/computer management, is intuitive. But they know how to use it and are used to that particular brand of buggy inconsistency.
Linux desktops have been high quality for a long time now. The reality is you, and others, measure quality as “how much is it like windows” or “how much of it is like macOS”. When that’s your metric, Linux will always come up short, just by definition.
Can I get a high performance Linux laptop with good battery life, fast graphics, that runs cool and silent?
What's high performance for you?
I can certainly get a Framework (Fedora and Ubuntu officially supported), throw my prefered Bluefin-Framework image in and get working
Battery life around 7 hours is the average I see reported, Fast/Silent will depend on the model, but I don't see the issue really Upgradability and easeness of battery replacement are a plus
I just picked framework because they were first to come to mind, but I think Dell has a nice Linux story, Tuxedo also comes to mind
These are the typical reviews I see around the Framework
https://community.frame.work/t/fw-16-review-the-good-the-bad...
Poor battery life, heavy, runs hot, poor build quality, bad speakers, and decent but not great graphics.
And yes, everything works. On bleeding edge 2 month old hardware.
I even use thunderbolt 4 to connect my external displays and peripherals. Not only does it work, but it’s pleasant. KDE has a settings panel for thunderbolt. I can even change my monitor brightness in KDE settings. No OSD required!
But wait, there’s more! I’m running 2 1440p monitors at 240hz and the system never even hiccups.
But wait, there’s more more! The battery settings are really advanced so I can change the power profile, maximum charge, everything.
The only thing I’m unsure about in your comment is “low latency audio”. It seems low latency to me, but I’m not an audio engineer.
This should not be an issue. I have hardware that varies a lot and I literally buy random wifi dongles for $1, $4, $5, Amazon, AliExpress, etc. and they have all just worked on first plugin. I can easily take my phone and tether it to my PC using USB-C and it appears in my Gnome network list and just starts using it for Internet.
> how well will it handle low latency audio
Pretty well you can use OBS to verify this. There are plenty of settings if you want to tune that.
> My graphics hardware?
Just ignore Nvidia and move on. Sure they might figure it out one day, I gave up a decade ago and I use Intel integrated or AMD dedicated for GPUs. Nvidia does "work" for most purposes but it will cause you a headache eventually and those are not worth $400 to me.
> How well will it handle power management?
I enjoy the basic controls that Gnome provides that give me a simple way to basically say "go all out" or "save some battery" etc. There are finer grain controls available and I have used commands in the past to push crappy hardware to it's limits before I chucked it (old Intel iGPUs)
> Can I get a high performance Linux laptop with good battery life, fast graphics, that runs cool and silent?
You can get ones that are specifically marketed for this purpose. Tuxedo is one that specializes in this and obviously System76 also do. These have a higher price point than a regular Dell system, which IMO is the better option in some ways. Dell sells more systems and has more users and it will "just work". They sold Linux systems for years and still do I believe.
Regarding "running silent" this is a gripe I have, not that it runs loud but some laptops have custom RGB crap and sometimes in Linux I don't have access to the extra functionality to customize my lighting or manually set my fans to ramp up etc. There are projects that aim to do this, but I have not looked into them beyond the most basic `fancontrol` built in command.
I think once you expand the scope to "most people" it might become impossible to say what the correct answer for that large of a group is. In the past their value add might have been more compelling and their feature lock not as draconian. It appears some people think that has changed over time.
Hard disagree. Audio mixing is not difficult[1]. The Linux kernel guys were right - it does not belong in the kernel. The userspace story however, has been a complete shitshow for decades. I think Pipewire mostly fixed that? Not sure, sometimes I still have to log out and back in to fix audio.
The funniest part? It's been working in the BSDs all along. I recommend reading the source of sndiod[1].
[1]: <https://cvsweb.openbsd.org/src/usr.bin/sndiod/>
What's even worse? Probably systemd. I try not to hold a strong opinion - I tolerate it, the way I tolerate traffic when taking a walk. The technical issue however is several orders of magnitude simpler - again, the BSDs come to mind, but you can also write a complete PID1 program in about 20 lines of straightforward C[2]. I don't mind the commands being unfamiliar (they're already all different in almost every OS family); it's that the entire package is dreadfully large in scope, opaque, and I find it more difficult to interact with than anything else in this landscape.
[2]: <https://ewontfix.com/14/>
However, it's worth noting that audio experts doing high grade mixing in production are using these systems quite effectively and have been for a long time. It's similar to Blender in that regard with it always having the "guts" of doing great things, but only the experts that knew the correct spells to cast were able to use it effectively before the UI/UX was improved with 2.x and later I believe.
Unfortunately, I still have to deal with macOS for work due to corporate policies.
I spent a day or so hacking around with kanata[0], which is a kernel level keyboard remapping tool, that lets you define keyboard mapping layers in a similar way you might with QMK firmware. When I press the 'super/win/cmd' it activates a layer which maps certain sequences to their control equivalents, so I can create tabs, close windows, copy and paste (and many more) like my macOS muscle memory wants to do. Other super key sequences (like Super-L for lock desktop or Super-Tab for window cycling) are unchanged. Furthermore, when I hit the control or meta/alt/option key, it activates a layer where Emacs editing keys are emulated using the Gnome equivalents. For example, C-a and C-e are mapped to home/end, etc.
The only problem is, this is not the behavior I want in terminals or in GNU/Emacs itself. So I installed a Gnome shell extension[1] that exports information about the active window state to a DBUS endpoint. That let me write a small python daemon (managed by a systemd user service) which wakes up whenever the active window changes. Based on this info, I send a message to the TCP server that kanata (also managed by a systemd user service) provides for remote control to switch to the appropriate layer.
After doing this, and tweaking my Gnome setup for another day or so, I am just as comfortable on my Linux machine as I was on my Mac. My main applications are Emacs, Firefox, Mattermost, Slack, ChatGPT, Discord, Kitty, and Steam. My Linux box was previously my Windows gaming box (don't get me started about frog boiling on Windows) and I'm amazed that I can play all my favorite titles (Manor Lords, Hell Let Loose, Foundation, Arma Reforager) on Linux with Proton.
[0]: https://github.com/jtroo/kanata
[1]: https://github.com/hseliger/window-calls-extended
I know it's mostly muscle memory, but macOS shortcuts just seem sane and consistent and that has been one of the biggest frustrations when trying to switch. I found toshy[0] which does something similar - did you try that? The goal is purely macOS key remappings in Linux, so a much smaller scope than kanata.
[0]: https://toshy.app
I have a Kinesis 360 keyboard, and my config[0] probably won't work for other keyboards, but it can give you a starting point for your own config.
[0]: https://gitlab.com/spudlyo/dotfiles/-/blob/master/kanata/.co...
If Linux, you'll have to be more specific, because users don't use Linux. They use Android, Ubuntu, Gnome, pop!os, redhat etc.
b) it might well be that the author really wishes they could draw a cartoon but can't, and resorted to AI to convey their aesthetic choices
I think your pessimism is unwarranted--AI illustrations do have their use.
Another thing I dislike is that it stores the whole message history on the device. It's nice to have at times, but I send a lot of photos, which adds up in storage over time. I pay for iCloud, and store my messages there. Why does my Mac need to hold every single photo I have ever sent?
[1] https://security.apple.com/blog/imessage-pq3/
Thankfully, that is also somehow the future of UI frameworks on all of their platforms!
Oh but you forgot about the “catch up” button they added 2 releases ago that takes you to the last unread message! …
… but only if said last message is within the N most recent messages, in the messages which are already “fetched” from local storage. If it’s more unread messages than that, the button is nowhere to be found.
Like they said “ok we can implement a catch up button but it’ll be hard to solve due to how we do paging.” “Ok we just won’t put the button on screen if we have to page then. Save the hard problem for the next release.” Then they just forgot about it.
- Ever since I've updated to the latest iOS 18, my watch complications(weather doodad) stop working randomly because they just lose the location services permission. Then in settings, the location services permission list acts like the weather app isn't installed.
- The new Mail app now automatically classifies your email, but still gives you the "All Mail" option. But the unread count badge on the app only works off of what they classify as your "Priority" mail. There's a setting to change that, so that it shows you the unread count of ALL mail, not just priority mail, but when you change that setting nothing changes. This is my biggest problem with new iOS.
- Keyboard sometimes doesn't get out the way any more when it should.
These are just off the top of my head. It used to be such a nice, polished experience. Their competition was just outclassed. Now, when my phone dies I'm going to have a good look at all the other options.
Depends on where you were seeing this of course, but this could very well be an app problem instead of a system problem.
Native UIKit/SwiftUI do a little bit of keyboard management for “free”, but there are many circumstances where it falls on the developer’s shoulders to do this. For cross platform frameworks, some do keyboard management others don’t even try. For web apps it’s a coin toss and depends on which of the gazillion ways the dev built their app.
It’s not actually that hard, usually just a matter of making sure that your scrolling content either resizes to match the keyboard-shrunken viewport or adding bottom padding equivalent to the height of the keyboard and then and adjusting scroll position accordingly, but it’s not unusual to see this partially or fully absent, especially on poorly built cheapest-bidder-contracted apps.
Of course SwiftUI gives you almost none of this control, forcing you to hope the magic automatic support works how you expect.
But then neither help you with any of the other interactions, like any background dimming you may want, or tapping away from the keyboard to dismiss. That has to be done manually.
I think they should just throw in the towel and duplicate settings. Meaning, we can turn off Siri learning from an app or from the Siri page. Or we can turn off banners from the app or the notifications page.
Had a few issues with iCloud syncing and data loss as well and what with being based in the UK and the general problems with geopolitics and the cloud I figured I'd try and get as much stuff out of iCloud as possible. Well there's not much advantage now. Most of it is in the ecosystem tie in, not the hardware. And on top of that the provisioned services such as Apple Music are just pain for me on a daily basis. My entire music catalogue disappeared in a puff of smoke when I was offline for nearly a week. The one thing I wanted it for!
So back to the PC. I ran out of disk space on the (soldered in SSD) Mac. I can't delete anything and macOS has leaked out about 20gb suddenly. I don't know what this is other than about 5 gig of it is Apple Intelligence despite telling it to fuck off. So it's late Friday afternoon and I need to get something done so I can have a clear weekend. I dig in the junk cupboard and find a couple of hard disks but no way of connecting them to the USB-C only Mac. Amazon solutions aren't available for delivery until Sunday. There upon I discovered the kids' "covid work PC" for when they were home studying. Despite the acceptable 16Gb of RAM it only had a meagre 256Gb disk in it. No worries. Opened it up and there's a hole for an SSD in it. It now has +500Gb SSD. Brilliant. On goes windows 11 LTSC. I'm back up running R in under an hour and have transferred all the data over.
I never went back. It feels better here. This thing is a swiss army knife. And extension of me. Not the other way round like on the Mac. The Mac feels like it feeds off me: both cash and energy. Apple need to fix that.
time machine?
That would be medium-term user. Long-term would be people like me that have been using it since 1984.
I have a collection of Macs going all the way back to 1984. Even the newest one hasn't been turned on in three years.
My daily driver is Windows Server 2016. But it has VMware Workstation so there are lots of virtual machines for my work, including Linux.
I am so tempted by the new M4s. Amazing piece of technology. So sad about the operating system though. Every year I say I'll wait for a quality Linux port.
There are three main choices and they are all compromised in their own way. You just need to figure out what is important to you and what isn't.
What you shouldn't do is take too much notice of posts like these, I've read through the whole thing and haven't had any of the issues mentioned. I've also not seen a mention of the issues I do have. HN has a negative tone, it seems we like to whinge.
I have to agree with this, System Settings seems very inconsistent (design) and has terrible information architecture / organization.