> They designed it, shipped it, lived with it for a year, then publicly changed their guidance on it. To me, that’s a lot more persuasive than simply saying, “Apple doesn’t do this.”
> Choosing not to do something you’ve never done is one thing.
> Choosing not to do something you’ve done is experience.
A more persuasive sales pitch for me to consider Apple products would be if the company had judgement not to ship such a hot mess of a sloppy product in the first place...
> A more persuasive sales pitch for me to consider Apple products would be if the company had judgement not to ship such a hot mess of a sloppy product in the first place...
When I read a sentence like this, I have to remind myself that this is Apple products we're talking about.
What other company do you feel surpasses Apple in terms of "judgement not to ship such a hot mess of a sloppy product in the first place"?
Is there any consumer electronics company today that ships products that are in any way less of a slop than Apple?
LLMs have a very strong voice, and that voice is “corporate hype press release”. So even fully human generated corporate crap sounds exactly like it. They both suck ass so they both deserve lambasting regardless.
It sounds like an LLM because LLMs were designed to sound like people. I get it, there’s a lot of tells and “quirks” that clearly mark out LLMs. But what makes an LLM obnoxious is not that it uses the same cliches we use all the time, but that it uses the same ones repetitively and sometimes in places where it doesn’t make sense.
But start paying attention to how you talk and write and you’ll realize you’re using those same cliches all the time. Heck I just did the “it’s not foo, it’s bar” cliche in the previous paragraph because it’s a useful language construct and it appears often in writing. An LLM writes like a budding author that learned “never use ‘blank said’” as a writing rule and goes to tortuous lengths to come up with more “clever” ways to say someone said something. But those clever ways are used all the time by very good writers too because they’re effective phrases. LLMs and hyper vigilance for their writing just means we’re more aware of the cliches we use
I agree with all of this historically ("it's not a bug, it's a feature" is the all time classic), but the problem is that LLMs have changed the rules of writing a little. Rhetorical flourishes of this ilk -- that might have seemed clever or witty before -- now feel very ick, especially when they come in that distinctive LLM metre.
TFA was an unusually extreme example, to the point where I do wonder if an LLM was involved in the writing thereof.
My inbox folders/labels used to be just text, and I'd have to read them to find the one I wanted to look at. I started adding emoji to the label names and suddenly I could scan the list far quicker to get to what I was looking for.
Granted, I had done this rather than the vendor deciding what icon fit - so I guess my brain had created neural pathways linking the image and the word.
Personally I think there's probably enough screen real estate to have both an icon and text in a menu entry, but menu search is generally more useful for anything not already an icon on the application's toolbar.
Yeah I appreciate their approach which optimizes for data density a lot more than competing desktops, but it could certainly feel cleaner while doing so.
I think mid-00s commercial OSes are a good target in this regard. OS X 10.2-10.9 and Windows XP-7 had a pretty good level of density while also feeling relatively clean and would be good to take some cues from.
Yeah with KDE it's all or nothing, either every menu item has an icon or you can turn off all of them.
This goes against the original argument which is that icons should only be used for universal idioms that can be recognized by the icon _only_. Otherwise it's just noise
It doesn't matter that it's JS, they have their own, performant interpreter for it and it's a limited subset of JS. And that's just one layer, the rest is actual compiled code, increasingly Rust.
I was skeptical, too, but it allows for some serious visual tweaking beyond what KDE allows without rewriting code.
Looking at all their design missteps and corrections in recent releases my only takeaway is that they are making it up as they go along just like the rest of us.
And on every misstep and their correction, the same people will be defending them. Like "icons on every option is so much more intuitive", just to argue the exact opposite a couple of years later.
> And on every misstep and their correction, the same people will be defending them.
That doesn't match my experience. Personally I've found Apple fans (not just Gruber and Nielsen but regular Apple users) to be the most vocally critical of their design missteps.
I think it might be somewhat generational. There was a time when Apple famously put a lot of effort into HCI research and experimentation, and prioritized it above so many other concerns. When Apple chose to do UI differently from everyone else, it was a strong indication that they made the decision based on empirical research and not just making it up as they go along like everyone else. A single menu bar at the top of the screen, for example, was chosen because they found that users could locate and click on them much faster than they could menus at the top of the window. I can’t remember exactly how much but I think it was something like 1/3 - 1/2 as much time.
IIRC Bill Atkinson said that it is at the top because it was easier to code. at any rate, it’s true that screen edges and corners are easy to click because the hit region becomes infinite in one or two directions
In this video Bill explains why he moved it. At 28 min he says: "…and if the window is near the bottom, how are you going to show all the choices? Well, I tried moving the menus to the top of the screen."
Certain designers may have hated Nielsen, but their users hated them, and they have more users hating them than Nielsen has designers hating him, and users matter much more than designers, so I think he came out way ahead.
Bruce Tognazzini is the OG GUI Guru of 80's user interface design!
Tog not just invented and implemented, but also deeply rationalized and documented a lot of great user interface techniques, like the "mile high menu bar", which partially exploits Fitts' Law (in the "up" direction), but made more sense on the original single small Mac screens. (While pie menus more fully exploit Fitts' law (in "all" directions") and they work great on large screens, giving you even more "leverage".)
>When the Macintosh was new, Bruce “Tog” Tognazzini wrote a column in Apple’s developer magazine on UI. In his column, people wrote in with lots of interesting UI design problems, which he discussed. These columns continue to this day on his web site. They’ve also been collected and embellished in a couple of great books, like Tog on Software Design, which is a lot of fun and a great introduction to UI design. (Tog on Interface was even better, but it’s out of print.)
>Tog invented the concept of the mile high menu bar to explain why the menu bar on the Macintosh, which is always glued to the top of the physical screen, is so much easier to use than menu bars on Windows, which appear inside each application window. When you want to point to the File menu on Windows, you have a target about half an inch wide and a quarter of an inch high to acquire. You must move and position the mouse fairly precisely in both the vertical and the horizontal dimensions.
>But on a Macintosh, you can slam the mouse up to the top of the screen, without regard to how high you slam it, and it will stop at the physical edge of the screen – the correct vertical position for using the menu. So, effectively, you have a target that is still half an inch wide, but a mile high. Now you only need to worry about positioning the cursor horizontally, not vertically, so the task of clicking on a menu item is that much easier.
>Based on this principle, Tog has a pop quiz: what are the five spots on the screen that are easiest to acquire (point to) with the mouse? The answer: all four corners of the screen (where you can literally slam the mouse over there in one fell swoop without any pointing at all), plus, the current position of the mouse, because it’s already there.
>The principle of the mile-high menu bar is fairly well known, but it must not be entirely obvious, because the Windows 95 team missed the point completely with the Start push button, sitting almost in the bottom left corner of the screen, but not exactly. In fact, it’s about 2 pixels away from the bottom and 2 pixels from the left of the screen. So, for the sake of a couple of pixels, Microsoft literally “snatches defeat from the jaws of victory”, Tog writes, and makes it that much harder to acquire the start button. It could have been a mile square, a...
I would really like to see an opensource or freely accessible version of HyperCard which does user interface right. Still annoyed by the rug-pull the LiveCode folks did for their opensource version and wish that the OpenXTalk folks would get some traction.
One of the most effective menu implementations ever for me was Altsys Virtuoso on NeXTstep's right-click (to pop-up the main menu at the mouse cursor), drag-down to Modify, then drag right to "Punch" or "Union" (whichever was the top command) --- it was a command I did a lot and it was almost gestural, kind of lke a well-done pie menu.
FWIW, the current Windows Start menu at lower-left, while not drawn to the corner now allows one to click the corner to open it, lower-right is "Show Desktop" and upper right closes the current maximized window --- only upper left is at the mercy of whatever app is currently running.
One criticism I have of HyperCard is that it wasn't "HyperCard All The Way Down", in that not all the rest of its dialogs and user interfaces and menus were made out of HyperCard stacks. It wasn't "metacircular", defined in terms of itself.
That's totally forgivable given that HyperCard was the first original iteration, and "merely" getting it finished in time to go on every Mac desktop was a spectacular achievement. But it doesn't excuse anything that came after it, and it doesn't excuse Apple for dropping it instead of evolving it.
CyberDog for OpenDoc was another interesting approach to a web browser, but they dropped that with OpenDoc too.
HyperLook, which was written in PostScript for the NeWS Window System, was deeply inspired by HyperCard, and addressed that as well as Atkinson's admitted limitation of being chained to a hard drive instead of networked. All of HyperLook's property sheets, dialogs, popups, help windows, menus, etc, were defined in terms of itself, editable and replaceable. Even the user interface editor itself was a replaceable component, so you could plug in different kinds of user interface editors, or remove it completely to deliver a non-editable runtime system (like SimCity).
The other thing that HyperLook added was another layer in the event delegation system: Button => Card => Background => Stack => Client. The client being what we would now call a "web server" or "distributed application", implementing what we now call the "AJAX" architecture.
I developed SimCity and HyperLook together, each one a forcing function for the other, and I'm taking the same approach now with reincarnating HyperTIES, Ben Shneiderman's pioneering hypertext system from the University of Maryland:
"It is an unworthy design objective to aim for anything less than trying to do to the Macintosh what the Macintosh did to the previous state of the art." -Bill Buxton
The "Natural" Language of Interaction: A Perspective on Non-Verbal Dialogues, INFOR: Canadian Journal of Operations Research and Information Processing, 26(4), 1988, pp. 428–438.
>Before going too much further, it is probably worth making a few comments about my approach and my examples. As will become pretty evident, the Apple Macin...
I would believe this, "single menu bar at the top, just update the values" is absolutely the solution you reach for when you're writing assembly by hand on an Apple II. It's hard to imagine but that UI element has been around since it started showing up in preview articles in ~1981
Exactly. There may have been a reason for the single menu bar back in 1984 when the Mac had a 9-inch screen. But it doesn't work very well on my 43-inch screen where the menu bar may be a few feet away from a small window in the lower right hand corner.
No, most people do not user their computer with 30" external screens. Perhaps developers at FAANGs do, but most people are mobile devices, follow by 13" or 15" laptop screens. Older people often use larger font sizes or browser zoom on top of that (or both).
I've seen internally collected telemetry. I don't remember the exact numbers, but for the application in question the size of the user's monitor was darn near a litmus test for whether they were one of our customers or a member of our development team.
It's funny trying to use an Office application on just a laptop's screen. You're left with a tiny slice of real estate because menus have gotten so bloated and don't adjust for the screen real estate you have. What's comfortable to work in on a 2x" monitor is not the same at <15".
I've seen this in webpages too. You load a page and think there's an error but the "login" button is surrounded by so much white space that it doesn't show up in the browser until you scroll down.
> Exactly. There may have been a reason for the single menu bar back in 1984 when the Mac had a 9-inch screen. But it doesn't work very well on my 43-inch screen where the menu bar may be a few feet away from a small window in the lower right hand corner.
That isn’t obvious. With a mouse/trackpad with a good acceleration curve, chances are you can move faster to that menu at the top than to a menu a few cm away because you don’t have to decelerate when approaching the menu bar at the top of the screen.
Last I checked, the mouse was a precision pointing device. It's stood the test of time because I can intuitively and precisely, without thinking, move to and click items anywhere on the screen.
Follow Apple's logic, reductio ad absurdum, and you'd have to put every single GUI element in the corners of the screen, because how else could you manage to quickly select them with a mouse? That OK button in the dialogue? Top right corner of the screen. Cancel? Top left. Quick, easy, and nobody could ever press the wrong button. It'll make it faster!
I'm not at all concerned with how easy it is to move the mouse to the menu. It's about looking at a window in one place, and having to look up and to the left by more than a foot to focus on the menu bar.
Resizing windows already requires moving the mouse with single-pixel precision. In comparison, hitting the menu bar is very easy.
As others have mentioned, having a target that the mouse automatically stops at, the screen edge, is far more ergonomic than a thin bar that requires more precision.
Also, these menus are very consistent in appearance, functionality, shortcuts, etc; moreso than my experience with Windows applications.
There are a lot of us that aren't making it up as we go. There is plenty of knowledge and experience on how to do these things and Apple used to be the king of it. For some reason they started throwing it all away with Liquid Glass.
Decades before Liquid Glass. They had begun throwing it all away by the time they got rid of scrollbars (and the rest of the industry followed like gormless lemmings).
Apple is/was regarded as the gods of design and ux. Does anyone know what it takes to get hired there as a designer now? Are the original designers with actual vision still there? I imagine they would need to really be conscious of not hiring average designers who just have a resume of top companies they have worked at but need the type of people steve jobs would have hired himself.
Easier said than done so it's not surprising they have atrophied in their former genius but im curious.
Steve Jobs himself talked about the bozo explosion that occurs when A players stop hiring A players. Was Alan Dye an A player? Certainly not to me. Who was the last Apple UX A player? Bas Ording? Mike Matas? Greg Christie? Imran Chaudri? The DNA is gone and the depletion started the moment Steve died and the infighting began leading to what we have today.
> they are making it up as they go along just like the rest of us.
"Everything around you that you call life was made up by people that are no smarter than you. And you can change it. And you can change it. You can influence it. You can build your own things that other people can use." —Steve Jobs
Hypocrisy would require commending Apple for icons everywhere and then celebrating their removal. That is not what happened. If anything has the homogeneity of hypocrisy, it’s this attitude that any commendation must be fanboying (while ignoring said commenters’ past criticism).
I think this and many other questionable UI decisions across the industry in recent years is the fallout of the old guard of UI designers having rotated out for younger replacements in the past decade with little transfer of knowledge occurring.
The lack of transfer I don't think is out of unwillingness from the old guard, but rather due to the greenhorns throwing everything that came before their time out without questioning the reasoning behind any of it (Chesterson's Fence) and assuming that whatever they came up with was superior and better fit for the modern era ("those old principles can't possibly hold, we have touch screens now!").
It's also been made worse by UI design roles having been taken by those with backgrounds in unrelated design disciplines, as well as sites like Dribbble which reward pretty mockups over working products. This leads to overemphasis on the visual aspect of the UI and not enough on its functional and usability aspects.
(1) Clickable elements no longer visually signifying that they're clickable. Words aren't underlined or written in a special font, sometimes the profile avatar is the clickable link to the profile but sometimes the name is, etc.
(2) No more clean, compact directory trees. We used to have high-contrast compact UI design to look through nested ontologies. Now it's rounded-edges card-based, often without trace lines, and barely ever compact. RIP
This is exhausting with web stuff. Every designer wants flat everything, not even so much as underlining links, let alone enough design to actually make it obvious what's a clickable button.
I think it's this cargo cult attempt to tap into what Apple had pre-2010 on the iPod and first iPhone. They're missing that the design leadership and attention to detail was unparalleled at Apple in that era, so they just flatten everything for a streamlined look and miss every signifier Apple put in its UIs in those days
>> This leads to overemphasis on the visual aspect of the UI and not enough on its functional and usability aspects.
Accessibility has been left in the dustbin as well. You go to all these sites extolling these beautiful, innovative sites and they're completely useless with a keyboard or screen reader.
Or if you have any sort of non-correctable visual impairment. I'm sighted but recently I had actually been thinking of learning better how to use a screen reader because I genuinely struggle to visually parse a lot of modern UI design. The big move toward flat designs, but with everyone choosing their own idiosyncratic special snowflake way of doing it, leaves very little visual redundancy for people to use as a fallback. Sad to hear that it might not be the solution either.
Also, holy hell, this obsession with huge swathes of negative space is a disaster for anyone who needs to zoom the page for readability. I'm here to see content, not an homage to the backrooms.
Yeah, same. It seems like the only times I actually feel disabled are when I’m using a web browser, and when I have to work in a project where somebody set the autoformatter’s line length up to 120 characters.
This weird love of negative space is particularly aggravating. Even when you're zoomed out you end up seeing nothing of note, then add in a 133% zoom and now you're reading a paragraph of text, scrolling horisontally and menus cover 75% of the viewport.
I am amazed at how UI design continuously come up with new reasons for additional margins and paddings. What used to be mere navigational links that were stacked vertically now have white space around the label, icons in front of the label, pill shape container around the clickable area, extra space between each link to give the container space to breath and a place to show the shadow effect.
Rightfully so. Your entire premise is that design should be made with accessibility in mind whereas in reality, accessibility tools should be built to handle those concerns and design should be free to do what it needs to do full stop. If that's being accessible that's fine, but trying to enforce accessibility standards is like saying that Levis needs to make pants for people without legs.
We let the wrong sort of designers do the design. The designers that simultaneously made changes under the guise of "UX" just sort of made it up as they went along without any basis in reality. That's how we got stuck with the hamburger menu.
I wish that were true. Unfortunately hardly a month goes by before I have to explain to someone that 1) those three little lines on top of each other aren't just a decoration but can be clicked, 2) that's where the settings live, 3) you've been in IT for 27 years but don't realize that's a thing you can click what a UX fail.
We can also lay some blame at Agile's feet. I've seen UX and UI designers genuinely struggle to fit their work into the sprint treadmill. 2 weeks' notice is not enough time to do any sort of serious usability design study.
The move toward fullstack developers, too. When everybody is a jack of all trades, nobody is a master of anything. Least of all UI design, which isn't worth thinking too much about because it's the least likely potential subject of pushback during code review.
I think the trick here is to not co-mingle the development of UI toolkits and the software built with them. Agile works fine if the UI framework itself is its own project with its own lifecycle, suitably long sprints, etc.
This goes against the pattern that's most often observed in the past decade of every product hand-rolling its UI from scratch, though.
I've worked at companies where to save a few thousands € to not buy a Qt license, they'd basically pay 50x more to have a bad copy of it that was also much slower due to using electron.
I think new UI designers might mistakenly think that they are in the fashion industry, when actually they are in the ergonomics engineering industry, or should be.
The amount of times I absolutely cringe from super lux UI which clearly a lot of time and energy went into prettifying at the absolute expense of usability and affordances. I love Apple TV but flip to any screen with multiple buttons and not being able to tell which button is selected. Or the Russian Roulette of Youtube trying to select any of the overlayed buttons on the actual video frame. I would 1000% trade shiny for more usablility in 99% of apps.
The Apple TV one drives me insane. The selected button is so subtly raised that it takes a lot to see. Usually I need to move it back and forth and look for the motion.
it really baffles me that Apple lets this happen with a design system/framework that has had years of battle testing. Clearly it's not working if today's apps are all insanely hard to use on a daily basis. Users that are power users. The system needs to protect us from app devs being able to ship stuff this hard to use. The remote is not helping of course.
To be its Car Play, amount of times I thought I have selected "Go", but it's actually "Back" that is selected. Drives me crazy. Partially fault of my car having shitty "touchpad" instead of a knob.
The YouTube UX is terrible. From watching a video in full-screen on an iPad to then going back to search (for the next video you want to see) takes three interactions (minimize, swipe down, then click on search).
I want to search, in a search bar, for the next video. This should not take three interactions.
YouTube is fine designing a UX where I am one click away from pulling up an AI-generated list of questions I may have about the video, though.
I don't like it either, but they are in the fashion industry.
Why did the label on my Fig Newtons get redesigned? It's not because it lacked in usability. It's because the company knows that customers respond to visual updates. If something looks the same year after year, lots of people will see it as old and tired. A redesign makes it feel modern and fresh. At least to enough people that it matters.
I had the same experience on a recent trip to The UPS Store. The color scheme was all different from the previous time I shipped a package. I commented to the clerk that they'd redecorated. He said they do it pretty often.
Software has a visual aspect to it, and the same forces are in effect. Part of a UI designer's job is to be on that treadmill. That's probably not what the job should be, but the people who pay their salary want it to be.
Well stated, but I will say, it is not impossible to do both.
Tantacrul is my favourite (and, actually, only) example of this. His work with MuseScore, with Audacity, his videos about them, and his takedowns of Sybellius etc. all opened my eyes to the world of designers who do things with hard-nosed analytical thinking, user studies, good principles, and make something useful. Even his little digressions about Paint3D and Ubuntu's Mobile OS are illustrative.
At least on the brick and mortar side of things it’s not so bad since the refresh also allows a maintenance cycle while the store is temporarily closed. And I can think of many examples (KMart especially) where the lack of store refreshes/deferred maintenance torpedoed the entire chain (Walmart and Target were not afraid to teardown or abandon outmoded locations).
Sure, Fig Newtons got redesigned because after seeing something over and over again it kinda blends, and design refresh and suddenly it catches your eye again.
No idea why UPS keeps redesigning other than signal "we still up to date with trends", its not like its rebranding going to make me want to ship anything that I don't already need to (and will ship with USPS unless its overnight in which case I will go with FedEx).
Interfaces get redesigned, in opinion, either because: a) just because (bad) b) certain things worked the way it worked only because it had to fit existing design and eventually old design just couldn't fit new feature(s) (can be good or bad, depending on execution)
I was so frustrated 1995-2005 that the designers I worked with had a print mindset. Now I'm frustrated that designers are no good at all.
It's no help at all that people expect Apple to be fashionable, like if you read a review of an "operating system" in Arstechnica you'd think it's all about little details on the scrollbar, not about filesystems and microkernels and namespaces and stuff.
A mere fashion industry would be fine, because you could just ignore that and stay old-fasioned. This fashion industry is coupled with enforced corporate dress code.
To be generous to the new generation and the people that hired them, the industry and its volume of designed products also grew a lot, super-linearly, in the last 20 years.
When that happens in an industry, the choices are mostly between having your experienced talent spread very more thinly or putting the brakes on unbridled expansion with some kind of licensure program or union.
This impacted all sides of digital products: engineering, graphic and experience design, product design, quality assurance, and even management. There was little option but to have the ranks fill in with amateurs, and those amateurs necessarily had a harder and harder time getting mentorship from people who'd been around the block. So they just earnestly started reinventing things, cargo culting, stumbling back into known disasters, etc
Presumably, they'll either figure things out eventually or some AI will help them learn what they missed for lack of mentorship.
> I think this and many other questionable UI decisions across the industry in recent years is the fallout of the old guard of UI designers having rotated out for younger replacements in the past decade with little transfer of knowledge occurring.
Not sure about that, given that John Gruber specifically lays the blame for a slew of recent shitty Apple design decisions on Alan Dye (cough liquid glass cough), and also the reversal of these shitty design decisions on changes made when Dye jumped ship for Meta. Gruber also says the Apple design team is much happier after Dye's departure.
Dye must be over 50, so I don't count him as a "younger replacement".
Alan Kay (of Smalltalk fame) has a similar, related explanation: computing is a pop culture with little regard for the past (https://link.springer.com/content/pdf/bbm:978-3-319-90008-7/...). There are thousands, maybe millions, of developers who have little knowledge of the history of our field. I’ve found this true even of graduates from highly-ranked computer science bachelor’s programs. Think of all of the current-day LLM researchers and engineers who have never heard of Minsky, McCarthy, Lisp machines, and other aspects of the history of AI before we learned that attention was all we needed.
It wouldn’t surprise me if Apple’s UI/UX teams are staffed with people who have never heard of Don Norman, Bruce Tognazzini, Bill Atkinson, Andy Hertzfeld, Susan Kare, Larry Tesler, and other pioneering figures who contributed to Apple’s legendary 1980s and 1990s software.
It’s a lack of onboarding and knowledge transfer, combined with an Eternal September of thousands new CS grads being constantly produced annually, who are sadly not being taught enough history of computing at their universities.
My dream as a CS professor is to help rectify this, at least amongst my students.
I had some hilarious experiences by telling people that they did well, now make the same thing again but make it better. I was quite surprised they put up with it many itterations and even more surprised that each round (with few exceptons) was actually better.
I concur, though I fear that as a discipline we haven't done ourselves any favours here in how we pass knowledge between generations.
In decades past, books and journals served as the primary mechanism for capturing and transferring CS and Software Engineering knowledge. Older knowledge was curated and remained easily accessible.
When I studied CS and engineering in the early 2000s, I learned a lot about the history and evolution of the field simply by browsing old books on the shelves of the university library. Thousands of them. I read back to the 60s and 70s, and I felt it really helped me have a broader perspective and evolution of the field.
That same university library has now transitioned to mostly online materials. The physical collection has been drastically pruned based on what is currently relevant and in demand.
The online collections don't have the same historical depth. And websites and online repositories come and go - think of the vast troves of knowledge that once existed on Usenet that have been lost to obscurity.
I also think the joy of browsing through topics you didn't know existed as given way to search-driven retrieval.
In our efforts to use computers to document what we know about computers, we have lost all that was good about not using computers for information management.
It’s hard to fully pass on UX knowledge. For one, a lot of software and hardware isn’t available, including recent software. You can’t use a version of Slack from 2019 or AIM from 2001, for instance.
But also, it’s hard to convey the full experience of actually using old software as a tool (or toy)—trying to maintain your company records in Lotus 1-2-3 or looking up something in Microsoft Encarta knowing that if you couldn’t answer your question you’d need to go to the library, or even trying to solve a puzzle in a game with no reference materials.
I’m sure this is true of other technology: You can learn to ride a horse or read about how telephones and telegrams worked, but can you really capture the experience of riding from NYC to Philadelphia in 1840 or trying to convey an urgent business message the same way in 1920?
"Computing is a pop culture with little regard for the past" this literally just clicked so many pieces in my brain all at once I had to close my eyes and sit with it.
Are they being taught any of significance? In school I only remember a blurb about Newton here or a blurb about Pascal there. Briefest of mentions about Unix at the start of a C class.
All the rest I had to dig up myself, from the Mythical Man Month to Edwin Black's book. Believe it was the Joel on Software blog who got me interested in those type of books.
It depends pretty much on the university, I was rather lucky, my nerdiness for strange OSes and programming languages comes from having had some great teachers, then again maybe up into the 90's there were another quality standards.
Alan Kay also said this about his famous "Point of View is worth 80 IQ points" aphorism, which I think poignantly explains why React costs a net of 160 IQ points compared to Svelte:
>I have been keeping to "80" IQ points in the saying (thus avoiding inflation). However very few people have noticed there is no sign bit in the aphorism, and that it works just as well for "poor perspectives being worth negative 80 IQ points" (and I think we could claim inflation of stupidity in many bad directions these days ...).
I love this quote! When I taught programming language paradigms at San Jose State as an adjunct, I introduced this quote as a motivating factor for learning different ways of programming, particularly introducing functional programming, logic programming, Smalltalk, and the Common Lisp Object System to senior-year students who coded mostly in Java, Python, and C (though this was during the COVID pandemic, right before the advent of vibe coding).
Here's the whole thread, started when I shared the "Folders" esolanguage article and hn discussion, which might be interesting for students and language hackers:
On 16 Oct 2026, Don Hopkins wrote to Owen Densmore, David Rosenthal, and Alan Kay:
Somebody took your filesystem-as-objects patent idea to the logical extreme!
I fear there is some really old prior art for several parts of it.
Best cheers
Alan
On 16 Oct 2026, Don Hopkins wrote:
Yes, it's Sun's US 5,187,786: "Method and apparatus for implementing a class hierarchy of objects in a hierarchical file system." Owen and David filed it in 1991, it was granted in 1993, and it expired in 2011, so anyone can use it now, including people writing esolangs about folders.
You're right about the prior art, and the idea came straight from your side of the family tree. Tom Stambaugh helped Owen work out how to adapt Smalltalk patterns to it, scribbling on napkins at a poolside table at the old Hyatt Palo Alto. In the patent, directories are classes and instances, PATH files encode the inheritance chain, and you send a message from the shell with something like "SEND aF001 methodA args".
It's the same trick Owen used for the NeWS object system in PostScript. Classes were dictionaries, and sending a message pushed the object's class chain onto the dictionary stack, so ordinary PostScript name lookup did the method lookup for free. NeWS 1.0 had single inheritance through a parent dictionary. TNT added ordered parent lists and multiple inheritance. The patent does the same thing with the shell path playing the dictionary stack: walk the directories, and the first match wins.
Owen's 1986 Monterey paper on object-oriented PostScript:
I've been running with it in MOOLLM, which leans more toward Self than Smalltalk. A directory is an object, its files are slots, and a "parents:" list delegates to other directories, in order, first match wins. It's prototypes rather than classes, and the LLM does the message dispatch by reading the directory. The filesystem is the object graph, so it's inspectable, diffable, and versioned in git for free:
I worked with all of the aforementioned people at various companies and cannot see myself working at any of the companies where this experience would be useful or could be transferred to others. Apple, when I left in 2006, was almost unbearable to me.
I don’t know if this means I am jaded, burnt out, overly cynical or what, but the thought of working at Apple, Google, Facebook or any other big tech company is anathema to me.
Steve Lemay worked with a bunch of us, including a very curmudgeonly Don Norman, but just looked what happened to Apple UX while he was there. Yes, he is now running the team and the small improvements may be a direct influence of his, but he sure formatted rolled by others the last twenty years.
Ahistoricality is a problem. On the other hand it might be that a small group that worked on some paradigm among the sea of many thinks they're the official forebearers of a large field - while in reality influence came from many places.
> It’s a lack of onboarding and knowledge transfer, ...
I have been trying to kindly forward what I have learnt and experienced over these years to new people we start to work with, and I have been called a wrinkly boomer who doesn't understand modern times more than enough times.
Of course they have come full-circle on some of the things they were disagreeing, but sometimes people have to crash and burn before understanding something.
I have no hard feelings for these people, but kindness goes a long way even if you don't agree on everything.
When the senior designers at Apple stared out, desktop UI was still being invented and they needed to be a kind of avant garde UI designer. They’re probably still hiring those people, but the job now is totally different, with desktop UI design being in a kind of maintenance mode.
Do you think the problem is that students are not taught about the history of computer science or is it because the kinds of students who study CS are not inclined to study history?
Reminds me of a funny encounter on a flight last year: just before the doors closed two young SWEs from the airline took the empty seats next to me on staff travel.
They'd recently entered the airline's graduate programme and when the girl told me she'd chosen the name Ada (Asian country where people generally pick their own international non-legal name) I was shocked to discover she had never heard of Ada Lovelace but "knew of" the programming language.
And I am a designer, not an engineer.
To be fair, the “younger replacements” have typically grown up surrounded with many more user interfaces than their predecessors, and the same is true for much of the audience they’re designing interfaces for.
Aesthetics aside, some of the specific guidance from the past is simply not as crucial as it used to be when most users were very new to smartphones or personal computers.
Back in the 2000s I would team up with people to put social applications in front of people and the one way you go could go big back then was carefully tune up every aspect of the funnel and get the viral branching ratio > 1 preferably >> 1 so you greased the skis and looked for every possible reason why people might not complete the process. It was life or death.
Around the time Facebook figured out what it's business model was platforms suppressed virality and people got used to the idea that you have to pay for traffic. You'd think that people would think "I'm paying $40 to acquire a customer, losing that customer to bad UI comes right out of my shareholder's pocketbook..." and somebody probably does, but a lot of people don't. Like zero interest rates and some people are just made of money and some people in business are lazy and, if their business is profitable, won't pursue cost reductions unless competition makes it life or death. Or something.
So I am appalled at so much bad software marketing that makes me think "is there anyone left who's had to tune up a marketing funnel?"
Like you see 200+ bug reports on Adobe's tracker about how Acrobat displays a total of 8 distracting UI elements to promote AI summaries the instant you open a document. And you can't clear them. And that's enough to make people see red.
But if you actually click on one of these stupid links you see a series of ads for things that are not AI features and that they don't make a serious effort to explain or sell. Then finally there's an ad for Acrobat Pro which maybe has the AI Summary but do I want to find out? I heard somebody somewhere did manage to try the AI Summary and of course it didn't work.
Like I could be annoyed by rapacious behavior but at least understand that people are motivated to make money for their shareholders. But to annoy everybody when they have Adobe CC subscriptions to lose and no clear path to convert? Where's the profit in that?
And that's totally ordinary. Like Meta solved so many tough problems to create the Meta Quest 3 and supposedly they are leaders in social software but they get the easy things wrong: they keep changing the interface that I'd use to "friend" people in games and make the process so terrible (at least the spammers who try to follow me on Instagram stole pictures of blonde women but nobody has a profile pic or the slightest bit of information to help decide "why would you follow me?") Supposedly Zuck wants to succeed at VR so bad... or is it just "we're #1 why try harder?"
One company whose site does get this is McMaster-Carr --- there used to be (maybe still is?) a great post on Quora about how the re-design for the site was realized and the very talented designer/architect who did it, but it's either gone, or I can't find it.
> "I'm paying $40 to acquire a customer, losing that customer to bad UI comes right out of my shareholder's pocketbook..." and somebody probably does, but a lot of people don't.
Yeah, this is insane. When I worked in marketing analytics I would point this out, and the most common response was that retention was a different team so we don't care about it.
Those who cannot remember the past are condemned to repeat it.
We came from CLI with white text on black. Now dark mode is all the rage for the last couple of years.
I know OLED and all but interfaces feel just like fashion now, to hell with affordances and improving how things actually work let's just make things pretty for that micro dopamine drip and change it up just for the sake of it because jazz hands looks new makes $$
I know this will never happen but it would be wonderful if designers were personally liable for all lost productivity of their changes. Designers move things around in some app or OS and millions of people have to stop their day-to-day work and figure out the new UI. Maybe you can attribute it to how many forests were burnt from the downstream effects of the changes.
Further, it's arguably become worse with touch screens in cars where, while you're driving, you get to discover a designer has moved a widget somewhere else and now your distracted. Self driving cars might solve that but even just walking around using your phone, every UI change is another distraction that might have consequences.
In Chesterton's fence analogy, the person who wants to tear a fence down is at least aware there is a fence there. I think a lot of modern UI is being done by people who aren't even aware that there was logic being used to make decisions and are just steamrolling over the fences without even being aware they existed.
In their defense, it's not like good UI has ever dominated very many spaces. It is very hard to come up with a concept of good UI concepts from a natural meandering through the shattered landscape of UI paradigms anyone computer literate ends up subjected to. And then it's even harder to convince a manager that it's worth so much as a moment's thought about such things. Not impossible. But not easy.
What about AI? Do you think that had no impact here? Certainly I've noticed all knowledge professions gain a sense of urgency, survival like psychosis-style thinking, and deterioration of quality and human thinking.
> I think this and many other questionable UI decisions across the industry in recent years is the fallout of the old guard of UI designers having rotated out for younger replacements in the past decade with little transfer of knowledge occurring.
Not just limited to UI decisions; many tech decisions were made the same way. I've said more than once, WRT to npm, that there were no adults in the room when they "designed" it, and npm is not the only tech stack that I look at and have the same thought.
The young'uns all cut their teeth on the web, where every project tends to reinvent the wheel (poorly).
It's funny (depressing) to me that there is now "web app" design language and a "native app" design language, and most new desktop apps are following the web because they are rebuilt in html and CSS.
Several years ago, I pointed out to a couple of UI designers that their design didn't take Fitts' Law into account. They looked at me like I was speaking ancient Sumerian. They had never heard of Fitts' Law, and they were supposedly degreed/credentialed/whatever UI designers. I was only a lowly software developer (albeit with a specialization in implmenting UI), and I knew what it was. I think it's only gotten worse.
The web ruined everything. Using a document platform for GUIs meant that every widget was reinvented over and over again until all conventional semantics were lost.
I am in what I feel is a unique position: I’ve been a dev and dev educator for 25 years, and I also have spent 15 years in HCI/UX/User Research.
I’ve spent countless hours facilitating usability tests and then building the software in response to them.
I 100% agree with the “dribblization” of usability engineering.
More so, I think the term “design” has become overloaded. A graphic designer who wants to actually get paid and eat goes into UX. But where does their joy come from? From the visual, not the usability.
This year I got hired at Nielsen Norman Group, helping train designers to understand the theories of the human mind and applying them to UI.
I am finding the range of knowledge and background amongst designers is massive, precisely because of the imprecision of the word “designer”.
If we called it “applied cognitive and behavioral psychology” we would get closer to the results we want and perhaps a smidge more respect for its necessity in the age of slop UI.
> If we called it “applied cognitive and behavioral psychology” we would get closer to the results we want and perhaps a smidge more respect for its necessity in the age of slop UI.
Yup, I have a psychology degree and we had courses in HCI. I ended up skipping them because the lecturer (who was great, and his notes were phenomenal) was just telling project management war stories, and I'd spent the few years previous in construction listening to the same stories in a different domain.
Now, 20 years later, I'm the one telling those stories.
>This leads to overemphasis on the visual aspect ....
For Apple, and specific to Apple, when they fired Scott Forstall, there is no one to protect the software UI guys. The 2012 reshuffle in the name of "collaboration" by Tim Cook saw Design team taking over everything design, including Software. 2013 saw iOS 7 and then they spend 3 years to walk back a lot of iOS 7 UI changes. In 2016 many of their old UI leads "retired". Having no old guards to defend their turf, giving Alan Dye more rooms for power grab.
Since then, if not earlier, despite the whole UI becoming more android and cluster mess, there have been little to zero major UI changes other than new skin.
And like Apple's software, that is 10 years of resume driven development where they just want to ship features after features.
I've been designing UI's professionally for 30 years and have seen a lot of changes over time, but I will echo what you said and give an additional reason: design education.
When I was getting my design degree, a large majority of what you learned was how to give and take a critique. It was central to everything you did. You were also taught things like "It's not your idea, it's an idea" that allowed you to critically think about solutions to problems without falling in love with your own designs.
This was fostered at the companies I worked for in the late 90's/early 2000's. We would have intense debates about designs, play devil's advocate (sometimes against our own designs!) in order to leave the room with the best ideas and solutions to problems. It was collaborative, energetic, peer reviewed, and we produced great designs for users.
This is not taught anymore. Critiques aren't given in school so as not to offend anyone. Much of UX Design has also transitioned to boot camps vs college design degrees which focus on pumping out students who can manipulate Figma, not students who understand the process and techniques of real design.
As such, professionally it's very hard to have real debates about design without someone getting offended or fighting for solutions that are sub par because it's "their" idea they can't let go of. The critical thinking and detail that used to go into design has been replaced with vibes.
And don't get me started on the fact that many students coming into the workforce simply don't have the design chops in the first place.
I don't quite understand. Icons by default obviously is stupid. But meaningful icons for actions such as copy and paste seems like a very good idea to me.
Personally I think there are not enough icons in MacOS right click menus for example. For me an icon makes it much easier to find the action I want to hse
I really hate the new trend to "clean up" UI everywhere which has the effect of not using screen space and forcing you to perform more actions in submenus.
I can't stand that keyboard shortcuts for menu items are the same color as the disabled menu item color. It's mixing two meanings for the same color in the same UI element. Despite this being the style for years, I still have a split-second interpretation of non-interactivity every time I see it.
When the Mac UI took over (some of) the world, many thought it was because of the icons, so they put icons all over their UIs.
That was of course not at all what made the MacOS so usable. The real principles were explained in detail in Apples UI guidelines. But hardly anyone read those, so the icons have been with us for 30 years.
What are the odds the next version of macOS reverts most of what happened in Tahoe? After having it forced on my work MacBook, I will not willingly install that trash UI on any device I own.
In asked myself that same question when Microsoft introduced Metro into Windows 8. Having been a Windows user at that point, it is the change that drove me to Mac OS X.
Fifteen years later, I am still waiting for them to undo the damage that was done.
It's awlays amusing hearing people bitch about Metro UI. It wasn't good, sure, but it was a product of the time.
And when people cried what 'oh noes i cant find a needed program in that awful fullscreen start menu uwu' I just show my Vista Start Menu list which took 4 FullHD screen heights on the small font settings.
And as people in the nearby comments say the problem here is what there is a lot of newcomers who just do shit - by mere lack of knowledge or worse by being KPI driven. Just like the last update of Mozilla Firefox which added ellipsised tabs. Like... for what? On the desktop?
I was actually a fan of Metro, coming from Windows Phone. I bought in early to Windows 8, including buying a Windows 8 Laptop (Lenovo X220), and a Surface RT.
The problem wasn't Metro itself -- it was the ham-fisted, halfhearted integration with the rest of Windows that proved irksome. Windows 8 was the version that introduced Settings alongside Control Panel -- 15 years later, the merge is almost complete.
They won't. They doubled down on Liquid Glass and its here to stay. Go take a look at their Apple Music rebrand. They honestly think it looks good! They really think they're cooking with their "we moved some acrylic blocks around our studio and thought real hard about it, aren't we great!" bullshit. It's so ugly.
Does anyone have a list of guidelines for good UI design that they would recommend?
As I have to design UI myself, I'd be quite interested. Personally, I do favor a mix in my UI: icons for obvious-or-frequently-used stuff (like Save, Undo, Help, Sort A-Z, etc, all of which usually have standard icons (except for Save now)) and plain text for things that can't be expressed by icon (like "Refactor XYZ").
and the high water mark of my graphical UI experience was when I used an NCR-3125 running Go Corp.'s PenPoint as a portable/note-taking/sketching device and a NeXT Cube w/ a Wacom ArtZ as my desktop.
Humans are very good at distinguishing shapes and colors. Visually distinct icons of various shapes help build muscle memory and quickly find the needed option.
One of the worst trends in icons were this insatiable desire to put different icons into uniform shapes, for example, this trend resulted in tragic replacement of beautiful varied and distinct Android icons that were introduced in Android 5 to the uglicity we have now.
On iOS, beautiful SFSymbol icons that floated beautifully on flat surfaces since iOS 7.0 were encased in absolutely similar circular 'liquid Glass' blobs. Sad.
I've never thought SF Symbols were beautiful. They are boring, ugly little bugs, hieroglyph crap that fade to blobs unless you're looking at them at 64x64 and above. Bereft of color, care, or soul, just crappy little vector squiggles. I hate them, and I hate that everyone copied them.
A company 1/100th the size used to design beautiful icons that clearly represented what they meant, and that company was Apple. Now I get SVG slop. SF Symbols can bite me.
That's also one reason to love those custom made hotchpotch UI. We don't see them to often anymore, but they had just the right functionality, and because the author was anything but a designer, the icons came from where ever, it was big colourful and buttons and icons sat in the same spot for a decade.
I don't think it helps if they're just random icons. Look at https://github.com/orgs/community/discussions . I went there a few days ago and immediately was distracted by all the random emoji icons scattered everywhere, and none of them provided me any useful information. I swear this is a new thing because it was just that noticeable. I had to put them in my adblock list.
Some very long time ago, in the early 00s, I read some UX study about icons. Ppl there replaced regular icons in app like MS word with random shapes, red square, blue circle, purple triangle, etc, and found out, that after initial learning period people find these icons just as fast as more meaningful ones. Recently, I tried finding that material, but couldn't find it. Still, it matches my experience with icons.
It would be interesting research to read and check their methodology if true.
Things are different today though, considering how many pieces of software an average person used in 2001 vs how many today. It's no longer just word and excel, updated once in a few years, so you can learn stuff by heart and not even care what icons they put on what
Most likely it was this: McDougall, S. J. P., de Bruijn, O., & Curry, M. B. (2000). Exploring the effects of icon characteristics on user performance: The role of icon concreteness, complexity, and distinctiveness
I think I stumbled upon it while reading that era's GNOME HIG, that had some very similar thoughts about icons..
> John Gruber wrote about this reversion back in June, including the fact that Apple’s new Human Interface Guidelines specifically call out Tahoe’s approach to menus as an example of what not to do on macOS.
Apple knew this was bad 20 years ago. They just don't bother to read their own documentation:
Apple Human Interface Guidelines (2008) [1] p. 170:
> If you do include icons in your menus, include them only for menu items for which they add significant value; don’t include them for every menu item. A menu that includes too many icons (or poorly designed ones) can appear cluttered and be hard to read.
Though the cog for settings is, like the three-dots commonly used to represent a menu itself, is only recognised as such because it has often been used as such. The first time(s) it was used it would not have been obvious to people. So the rule isn't that a good icon should be obvious. A good icon is either immediately obvious or consistently used.
The common save icon is a good example of this in both cases: to those of us who have been around a while it is obvious because it looks like a floppy disk, to those who never used floppy disks it is obvious because that is what save icons (almost) always are.
I don't love the insinuation that people making mistakes (especially mistakes like failing to read and retain one of probably hundreds of thousands of pieces of UI guidance) is somehow a character flaw. "Apple" isn't a person, and just because it's written down doesn't mean it gets memorized by every individual on every team.
Apple isn't a person. It doesn't need to memorize. Its a Apple failing because no process caught it, and because anything coming out of Apple that is bad is a failing by Apple.
“They don’t bother to read their own documentation” to me reads more like an indictment of people’s priorities than criticizing a business practice. It makes it sound like even though the business process says to do X, lazy employees did Y instead. If that wasn’t the intent then the wording could have been more carefully chosen. But it’s just a small gripe :)
> Apple knew this was bad 20 years ago. They just don't bother to read their own documentation
This isn't really a case of someone overlooking previous guidance, but a symptom of an org where platform consistency is being prioritized over tradition. This means changes now cater to the lowest common denominator (iOS). Lots of recent updates this, like the new app icons, the macOS 26 implementation of liquid glass, etc. The priority is making macOS easier to adopt for people who's first devices were iPads.
It's unfortunate. Although what I find strange are some of the very obtuse patterns they refuse to change. It took them forever to add window snapping to macOS and Finder still doesn't have a cut command.
The only time I think you ever need an icon beside text is if the exact same feature is found elsewhere as an icon only. And even then ask yourself if it can be avoided.
I think icon in every menu is fine. The problem is choosing the wrong icon. The icon could just be text, you know, Like [TU]:Tuesday.
English is not my native language, so i was having a very hard time to learn how to use computer, with icon, my 8 Years old brain can kind of infer what is its meaning.
221 comments
[ 2.9 ms ] story [ 13.7 ms ] thread> They designed it, shipped it, lived with it for a year, then publicly changed their guidance on it. To me, that’s a lot more persuasive than simply saying, “Apple doesn’t do this.”
> Choosing not to do something you’ve never done is one thing.
> Choosing not to do something you’ve done is experience.
A more persuasive sales pitch for me to consider Apple products would be if the company had judgement not to ship such a hot mess of a sloppy product in the first place...
When I read a sentence like this, I have to remind myself that this is Apple products we're talking about.
What other company do you feel surpasses Apple in terms of "judgement not to ship such a hot mess of a sloppy product in the first place"?
Is there any consumer electronics company today that ships products that are in any way less of a slop than Apple?
Alan Dye really did a number on that company and set back UI design by one or two decades, favoring graphic design over human interface design.
> > Choosing not to do something you’ve never done is one thing.
> > Choosing not to do something you’ve done is experience.
Gods, this sounds so LLM :(
But start paying attention to how you talk and write and you’ll realize you’re using those same cliches all the time. Heck I just did the “it’s not foo, it’s bar” cliche in the previous paragraph because it’s a useful language construct and it appears often in writing. An LLM writes like a budding author that learned “never use ‘blank said’” as a writing rule and goes to tortuous lengths to come up with more “clever” ways to say someone said something. But those clever ways are used all the time by very good writers too because they’re effective phrases. LLMs and hyper vigilance for their writing just means we’re more aware of the cliches we use
TFA was an unusually extreme example, to the point where I do wonder if an LLM was involved in the writing thereof.
2. Click on the hamburger menu in the top right
3. ???
Icons can genuinely help convey meaning/understanding. They can also be clutter. It comes down to how you use them.
Granted, I had done this rather than the vendor deciding what icon fit - so I guess my brain had created neural pathways linking the image and the word.
Personally I think there's probably enough screen real estate to have both an icon and text in a menu entry, but menu search is generally more useful for anything not already an icon on the application's toolbar.
I think mid-00s commercial OSes are a good target in this regard. OS X 10.2-10.9 and Windows XP-7 had a pretty good level of density while also feeling relatively clean and would be good to take some cues from.
This goes against the original argument which is that icons should only be used for universal idioms that can be recognized by the icon _only_. Otherwise it's just noise
I've been hoping Cosmic DE improves enough to be usable.
I was skeptical, too, but it allows for some serious visual tweaking beyond what KDE allows without rewriting code.
Looking at all their design missteps and corrections in recent releases my only takeaway is that they are making it up as they go along just like the rest of us.
That doesn't match my experience. Personally I've found Apple fans (not just Gruber and Nielsen but regular Apple users) to be the most vocally critical of their design missteps.
You don’t have to make it up, it was only one year ago. All their blog posts are still visible.
But that was in a different millennium.
https://milos.fyi/blog/fitts-law-why-the-menu-bar-is-at-the-...
https://www.cs.cmu.edu/~bam/uicourse/05440inter2019/Bill-Atk...
https://www.cs.cmu.edu/~bam/uicourse/05440inter2019/Don-Hopk...
More on the "Mile High Menu Bar":
https://news.ycombinator.com/item?id=44125835
Certain designers may have hated Nielsen, but their users hated them, and they have more users hating them than Nielsen has designers hating him, and users matter much more than designers, so I think he came out way ahead.
Bruce Tognazzini is the OG GUI Guru of 80's user interface design!
https://asktog.com/atc/about-bruce-tognazzini/
Tog not just invented and implemented, but also deeply rationalized and documented a lot of great user interface techniques, like the "mile high menu bar", which partially exploits Fitts' Law (in the "up" direction), but made more sense on the original single small Mac screens. (While pie menus more fully exploit Fitts' law (in "all" directions") and they work great on large screens, giving you even more "leverage".)
https://www.joelonsoftware.com/2000/04/27/designing-for-peop...
>When the Macintosh was new, Bruce “Tog” Tognazzini wrote a column in Apple’s developer magazine on UI. In his column, people wrote in with lots of interesting UI design problems, which he discussed. These columns continue to this day on his web site. They’ve also been collected and embellished in a couple of great books, like Tog on Software Design, which is a lot of fun and a great introduction to UI design. (Tog on Interface was even better, but it’s out of print.)
>Tog invented the concept of the mile high menu bar to explain why the menu bar on the Macintosh, which is always glued to the top of the physical screen, is so much easier to use than menu bars on Windows, which appear inside each application window. When you want to point to the File menu on Windows, you have a target about half an inch wide and a quarter of an inch high to acquire. You must move and position the mouse fairly precisely in both the vertical and the horizontal dimensions.
>But on a Macintosh, you can slam the mouse up to the top of the screen, without regard to how high you slam it, and it will stop at the physical edge of the screen – the correct vertical position for using the menu. So, effectively, you have a target that is still half an inch wide, but a mile high. Now you only need to worry about positioning the cursor horizontally, not vertically, so the task of clicking on a menu item is that much easier.
>Based on this principle, Tog has a pop quiz: what are the five spots on the screen that are easiest to acquire (point to) with the mouse? The answer: all four corners of the screen (where you can literally slam the mouse over there in one fell swoop without any pointing at all), plus, the current position of the mouse, because it’s already there.
>The principle of the mile-high menu bar is fairly well known, but it must not be entirely obvious, because the Windows 95 team missed the point completely with the Start push button, sitting almost in the bottom left corner of the screen, but not exactly. In fact, it’s about 2 pixels away from the bottom and 2 pixels from the left of the screen. So, for the sake of a couple of pixels, Microsoft literally “snatches defeat from the jaws of victory”, Tog writes, and makes it that much harder to acquire the start button. It could have been a mile square, a...
One of the most effective menu implementations ever for me was Altsys Virtuoso on NeXTstep's right-click (to pop-up the main menu at the mouse cursor), drag-down to Modify, then drag right to "Punch" or "Union" (whichever was the top command) --- it was a command I did a lot and it was almost gestural, kind of lke a well-done pie menu.
FWIW, the current Windows Start menu at lower-left, while not drawn to the corner now allows one to click the corner to open it, lower-right is "Show Desktop" and upper right closes the current maximized window --- only upper left is at the mercy of whatever app is currently running.
One criticism I have of HyperCard is that it wasn't "HyperCard All The Way Down", in that not all the rest of its dialogs and user interfaces and menus were made out of HyperCard stacks. It wasn't "metacircular", defined in terms of itself.
That's totally forgivable given that HyperCard was the first original iteration, and "merely" getting it finished in time to go on every Mac desktop was a spectacular achievement. But it doesn't excuse anything that came after it, and it doesn't excuse Apple for dropping it instead of evolving it.
CyberDog for OpenDoc was another interesting approach to a web browser, but they dropped that with OpenDoc too.
HyperLook, which was written in PostScript for the NeWS Window System, was deeply inspired by HyperCard, and addressed that as well as Atkinson's admitted limitation of being chained to a hard drive instead of networked. All of HyperLook's property sheets, dialogs, popups, help windows, menus, etc, were defined in terms of itself, editable and replaceable. Even the user interface editor itself was a replaceable component, so you could plug in different kinds of user interface editors, or remove it completely to deliver a non-editable runtime system (like SimCity).
The other thing that HyperLook added was another layer in the event delegation system: Button => Card => Background => Stack => Client. The client being what we would now call a "web server" or "distributed application", implementing what we now call the "AJAX" architecture.
HyperLook (nee HyperNews (nee GoodNews)):
https://donhopkins.medium.com/hyperlook-nee-hypernews-nee-go...
I developed SimCity and HyperLook together, each one a forcing function for the other, and I'm taking the same approach now with reincarnating HyperTIES, Ben Shneiderman's pioneering hypertext system from the University of Maryland:
https://github.com/SimHacker/Hyperties/blob/main/docs/README...
https://hyperties.org/
I recently wrote this about cloning Photoshop, quoting Alan Kay about old ideas that hang on, and Bill Buxton about unworthy design objectives.
https://news.ycombinator.com/item?id=50052665
Cloning Photoshop is such an underwhelmingly unoriginal and unambitious design goal, like photocopying a Roy Lichtenstein painting, or cloning MS-DOS.
Treating Photoshop's interface as the destination squanders the opportunity to rethink how people create images.
"Things seem to hang on in computing just because they work a little bit." -Alan Kay
From a personal email about MVC and Morphic, published here:
https://news.ycombinator.com/item?id=8841428
"It is an unworthy design objective to aim for anything less than trying to do to the Macintosh what the Macintosh did to the previous state of the art." -Bill Buxton
The "Natural" Language of Interaction: A Perspective on Non-Verbal Dialogues, INFOR: Canadian Journal of Operations Research and Information Processing, 26(4), 1988, pp. 428–438.
https://www.billbuxton.com/natural.html
>THE MACINTOSH AS VICTIM
>Before going too much further, it is probably worth making a few comments about my approach and my examples. As will become pretty evident, the Apple Macin...
Exactly. There may have been a reason for the single menu bar back in 1984 when the Mac had a 9-inch screen. But it doesn't work very well on my 43-inch screen where the menu bar may be a few feet away from a small window in the lower right hand corner.
I've seen this in webpages too. You load a page and think there's an error but the "login" button is surrounded by so much white space that it doesn't show up in the browser until you scroll down.
That isn’t obvious. With a mouse/trackpad with a good acceleration curve, chances are you can move faster to that menu at the top than to a menu a few cm away because you don’t have to decelerate when approaching the menu bar at the top of the screen.
Follow Apple's logic, reductio ad absurdum, and you'd have to put every single GUI element in the corners of the screen, because how else could you manage to quickly select them with a mouse? That OK button in the dialogue? Top right corner of the screen. Cancel? Top left. Quick, easy, and nobody could ever press the wrong button. It'll make it faster!
Resizing windows already requires moving the mouse with single-pixel precision. In comparison, hitting the menu bar is very easy.
Also, these menus are very consistent in appearance, functionality, shortcuts, etc; moreso than my experience with Windows applications.
Easier said than done so it's not surprising they have atrophied in their former genius but im curious.
Best known from The Bible, Steve Jobs, Donald Trump and Adolf Hitler.
"Everything around you that you call life was made up by people that are no smarter than you. And you can change it. And you can change it. You can influence it. You can build your own things that other people can use." —Steve Jobs
Full clip slightly longer, worth watching.
https://youtube.com/watch?v=kYfNvmF0Bqw
It's not perfect here but having an ergonomic design and replacing it with a substantially worse one is rather unheard of in this world.
Hypocrisy would require commending Apple for icons everywhere and then celebrating their removal. That is not what happened. If anything has the homogeneity of hypocrisy, it’s this attitude that any commendation must be fanboying (while ignoring said commenters’ past criticism).
Your assumptions only work until they don't. It's not like we live in a static situation. What didn't work yesterday might work today
The lack of transfer I don't think is out of unwillingness from the old guard, but rather due to the greenhorns throwing everything that came before their time out without questioning the reasoning behind any of it (Chesterson's Fence) and assuming that whatever they came up with was superior and better fit for the modern era ("those old principles can't possibly hold, we have touch screens now!").
It's also been made worse by UI design roles having been taken by those with backgrounds in unrelated design disciplines, as well as sites like Dribbble which reward pretty mockups over working products. This leads to overemphasis on the visual aspect of the UI and not enough on its functional and usability aspects.
(1) Clickable elements no longer visually signifying that they're clickable. Words aren't underlined or written in a special font, sometimes the profile avatar is the clickable link to the profile but sometimes the name is, etc.
(2) No more clean, compact directory trees. We used to have high-contrast compact UI design to look through nested ontologies. Now it's rounded-edges card-based, often without trace lines, and barely ever compact. RIP
Accessibility has been left in the dustbin as well. You go to all these sites extolling these beautiful, innovative sites and they're completely useless with a keyboard or screen reader.
Also, holy hell, this obsession with huge swathes of negative space is a disaster for anyone who needs to zoom the page for readability. I'm here to see content, not an homage to the backrooms.
The move toward fullstack developers, too. When everybody is a jack of all trades, nobody is a master of anything. Least of all UI design, which isn't worth thinking too much about because it's the least likely potential subject of pushback during code review.
This goes against the pattern that's most often observed in the past decade of every product hand-rolling its UI from scratch, though.
I want to search, in a search bar, for the next video. This should not take three interactions.
YouTube is fine designing a UX where I am one click away from pulling up an AI-generated list of questions I may have about the video, though.
Why did the label on my Fig Newtons get redesigned? It's not because it lacked in usability. It's because the company knows that customers respond to visual updates. If something looks the same year after year, lots of people will see it as old and tired. A redesign makes it feel modern and fresh. At least to enough people that it matters.
I had the same experience on a recent trip to The UPS Store. The color scheme was all different from the previous time I shipped a package. I commented to the clerk that they'd redecorated. He said they do it pretty often.
Software has a visual aspect to it, and the same forces are in effect. Part of a UI designer's job is to be on that treadmill. That's probably not what the job should be, but the people who pay their salary want it to be.
Tantacrul is my favourite (and, actually, only) example of this. His work with MuseScore, with Audacity, his videos about them, and his takedowns of Sybellius etc. all opened my eyes to the world of designers who do things with hard-nosed analytical thinking, user studies, good principles, and make something useful. Even his little digressions about Paint3D and Ubuntu's Mobile OS are illustrative.
No idea why UPS keeps redesigning other than signal "we still up to date with trends", its not like its rebranding going to make me want to ship anything that I don't already need to (and will ship with USPS unless its overnight in which case I will go with FedEx).
Interfaces get redesigned, in opinion, either because: a) just because (bad) b) certain things worked the way it worked only because it had to fit existing design and eventually old design just couldn't fit new feature(s) (can be good or bad, depending on execution)
It's no help at all that people expect Apple to be fashionable, like if you read a review of an "operating system" in Arstechnica you'd think it's all about little details on the scrollbar, not about filesystems and microkernels and namespaces and stuff.
Bad Dye Job:
https://daringfireball.net/2025/12/bad_dye_job
https://news.ycombinator.com/item?id=46191194
When that happens in an industry, the choices are mostly between having your experienced talent spread very more thinly or putting the brakes on unbridled expansion with some kind of licensure program or union.
This impacted all sides of digital products: engineering, graphic and experience design, product design, quality assurance, and even management. There was little option but to have the ranks fill in with amateurs, and those amateurs necessarily had a harder and harder time getting mentorship from people who'd been around the block. So they just earnestly started reinventing things, cargo culting, stumbling back into known disasters, etc
Presumably, they'll either figure things out eventually or some AI will help them learn what they missed for lack of mentorship.
Not sure about that, given that John Gruber specifically lays the blame for a slew of recent shitty Apple design decisions on Alan Dye (cough liquid glass cough), and also the reversal of these shitty design decisions on changes made when Dye jumped ship for Meta. Gruber also says the Apple design team is much happier after Dye's departure.
Dye must be over 50, so I don't count him as a "younger replacement".
It wouldn’t surprise me if Apple’s UI/UX teams are staffed with people who have never heard of Don Norman, Bruce Tognazzini, Bill Atkinson, Andy Hertzfeld, Susan Kare, Larry Tesler, and other pioneering figures who contributed to Apple’s legendary 1980s and 1990s software.
It’s a lack of onboarding and knowledge transfer, combined with an Eternal September of thousands new CS grads being constantly produced annually, who are sadly not being taught enough history of computing at their universities.
My dream as a CS professor is to help rectify this, at least amongst my students.
In decades past, books and journals served as the primary mechanism for capturing and transferring CS and Software Engineering knowledge. Older knowledge was curated and remained easily accessible.
When I studied CS and engineering in the early 2000s, I learned a lot about the history and evolution of the field simply by browsing old books on the shelves of the university library. Thousands of them. I read back to the 60s and 70s, and I felt it really helped me have a broader perspective and evolution of the field.
That same university library has now transitioned to mostly online materials. The physical collection has been drastically pruned based on what is currently relevant and in demand.
The online collections don't have the same historical depth. And websites and online repositories come and go - think of the vast troves of knowledge that once existed on Usenet that have been lost to obscurity.
I also think the joy of browsing through topics you didn't know existed as given way to search-driven retrieval.
In our efforts to use computers to document what we know about computers, we have lost all that was good about not using computers for information management.
But also, it’s hard to convey the full experience of actually using old software as a tool (or toy)—trying to maintain your company records in Lotus 1-2-3 or looking up something in Microsoft Encarta knowing that if you couldn’t answer your question you’d need to go to the library, or even trying to solve a puzzle in a game with no reference materials.
I’m sure this is true of other technology: You can learn to ride a horse or read about how telephones and telegrams worked, but can you really capture the experience of riding from NYC to Philadelphia in 1840 or trying to convey an urgent business message the same way in 1920?
Are they being taught any of significance? In school I only remember a blurb about Newton here or a blurb about Pascal there. Briefest of mentions about Unix at the start of a C class.
All the rest I had to dig up myself, from the Mythical Man Month to Edwin Black's book. Believe it was the Joel on Software blog who got me interested in those type of books.
https://news.ycombinator.com/item?id=8841428
Alan Kay also said this about his famous "Point of View is worth 80 IQ points" aphorism, which I think poignantly explains why React costs a net of 160 IQ points compared to Svelte:
>I have been keeping to "80" IQ points in the saying (thus avoiding inflation). However very few people have noticed there is no sign bit in the aphorism, and that it works just as well for "poor perspectives being worth negative 80 IQ points" (and I think we could claim inflation of stupidity in many bad directions these days ...).
On 16 Oct 2026, Don Hopkins wrote to Owen Densmore, David Rosenthal, and Alan Kay:
Somebody took your filesystem-as-objects patent idea to the logical extreme!
https://danieltemkin.com/esolangs/folders
https://news.ycombinator.com/item?id=50114342
-Don
On 16 Oct 2026, Alan Kay wrote:
There's a patent for this idea?
I fear there is some really old prior art for several parts of it.
Best cheers
Alan
On 16 Oct 2026, Don Hopkins wrote:
Yes, it's Sun's US 5,187,786: "Method and apparatus for implementing a class hierarchy of objects in a hierarchical file system." Owen and David filed it in 1991, it was granted in 1993, and it expired in 2011, so anyone can use it now, including people writing esolangs about folders.
https://patents.google.com/patent/US5187786A/en
You're right about the prior art, and the idea came straight from your side of the family tree. Tom Stambaugh helped Owen work out how to adapt Smalltalk patterns to it, scribbling on napkins at a poolside table at the old Hyatt Palo Alto. In the patent, directories are classes and instances, PATH files encode the inheritance chain, and you send a message from the shell with something like "SEND aF001 methodA args".
It's the same trick Owen used for the NeWS object system in PostScript. Classes were dictionaries, and sending a message pushed the object's class chain onto the dictionary stack, so ordinary PostScript name lookup did the method lookup for free. NeWS 1.0 had single inheritance through a parent dictionary. TNT added ordered parent lists and multiple inheritance. The patent does the same thing with the shell path playing the dictionary stack: walk the directories, and the first match wins.
Owen's 1986 Monterey paper on object-oriented PostScript:
https://donhopkins.com/home/monterey86.pdf
The NeWS Book, chapter 6:
https://donhopkins.com/home/The_NeWS_Book_1989.pdf
I've been running with it in MOOLLM, which leans more toward Self than Smalltalk. A directory is an object, its files are slots, and a "parents:" list delegates to other directories, in order, first match wins. It's prototypes rather than classes, and the LLM does the message dispatch by reading the directory. The filesystem is the object graph, so it's inspectable, diffable, and versioned in git for free:
https://github.com/SimHacker/moollm/blob/main/kernel/DIRECTO...
https://github.com/SimHacker/moollm/blob/main/skills/file-sy...
The lineage from Smalltalk and Self through NeWS, TNT, the patent, HyperLook, and The Sims to MOOLLM is written up here:
https://github.com/SimHacker/moollm/blob/main/designs/object...
I don’t know if this means I am jaded, burnt out, overly cynical or what, but the thought of working at Apple, Google, Facebook or any other big tech company is anathema to me.
Steve Lemay worked with a bunch of us, including a very curmudgeonly Don Norman, but just looked what happened to Apple UX while he was there. Yes, he is now running the team and the small improvements may be a direct influence of his, but he sure formatted rolled by others the last twenty years.
I have been trying to kindly forward what I have learnt and experienced over these years to new people we start to work with, and I have been called a wrinkly boomer who doesn't understand modern times more than enough times.
Of course they have come full-circle on some of the things they were disagreeing, but sometimes people have to crash and burn before understanding something.
I have no hard feelings for these people, but kindness goes a long way even if you don't agree on everything.
Aesthetics aside, some of the specific guidance from the past is simply not as crucial as it used to be when most users were very new to smartphones or personal computers.
Around the time Facebook figured out what it's business model was platforms suppressed virality and people got used to the idea that you have to pay for traffic. You'd think that people would think "I'm paying $40 to acquire a customer, losing that customer to bad UI comes right out of my shareholder's pocketbook..." and somebody probably does, but a lot of people don't. Like zero interest rates and some people are just made of money and some people in business are lazy and, if their business is profitable, won't pursue cost reductions unless competition makes it life or death. Or something.
So I am appalled at so much bad software marketing that makes me think "is there anyone left who's had to tune up a marketing funnel?"
Like you see 200+ bug reports on Adobe's tracker about how Acrobat displays a total of 8 distracting UI elements to promote AI summaries the instant you open a document. And you can't clear them. And that's enough to make people see red.
But if you actually click on one of these stupid links you see a series of ads for things that are not AI features and that they don't make a serious effort to explain or sell. Then finally there's an ad for Acrobat Pro which maybe has the AI Summary but do I want to find out? I heard somebody somewhere did manage to try the AI Summary and of course it didn't work.
Like I could be annoyed by rapacious behavior but at least understand that people are motivated to make money for their shareholders. But to annoy everybody when they have Adobe CC subscriptions to lose and no clear path to convert? Where's the profit in that?
And that's totally ordinary. Like Meta solved so many tough problems to create the Meta Quest 3 and supposedly they are leaders in social software but they get the easy things wrong: they keep changing the interface that I'd use to "friend" people in games and make the process so terrible (at least the spammers who try to follow me on Instagram stole pictures of blonde women but nobody has a profile pic or the slightest bit of information to help decide "why would you follow me?") Supposedly Zuck wants to succeed at VR so bad... or is it just "we're #1 why try harder?"
Yeah, this is insane. When I worked in marketing analytics I would point this out, and the most common response was that retention was a different team so we don't care about it.
That being said, this is an extremely common problem across basically every industry, and perhaps it's inherent to how companies are currently run.
We came from CLI with white text on black. Now dark mode is all the rage for the last couple of years.
I know OLED and all but interfaces feel just like fashion now, to hell with affordances and improving how things actually work let's just make things pretty for that micro dopamine drip and change it up just for the sake of it because jazz hands looks new makes $$
Further, it's arguably become worse with touch screens in cars where, while you're driving, you get to discover a designer has moved a widget somewhere else and now your distracted. Self driving cars might solve that but even just walking around using your phone, every UI change is another distraction that might have consequences.
In their defense, it's not like good UI has ever dominated very many spaces. It is very hard to come up with a concept of good UI concepts from a natural meandering through the shattered landscape of UI paradigms anyone computer literate ends up subjected to. And then it's even harder to convince a manager that it's worth so much as a moment's thought about such things. Not impossible. But not easy.
Not just limited to UI decisions; many tech decisions were made the same way. I've said more than once, WRT to npm, that there were no adults in the room when they "designed" it, and npm is not the only tech stack that I look at and have the same thought.
It's funny (depressing) to me that there is now "web app" design language and a "native app" design language, and most new desktop apps are following the web because they are rebuilt in html and CSS.
I’ve spent countless hours facilitating usability tests and then building the software in response to them.
I 100% agree with the “dribblization” of usability engineering.
More so, I think the term “design” has become overloaded. A graphic designer who wants to actually get paid and eat goes into UX. But where does their joy come from? From the visual, not the usability.
This year I got hired at Nielsen Norman Group, helping train designers to understand the theories of the human mind and applying them to UI.
I am finding the range of knowledge and background amongst designers is massive, precisely because of the imprecision of the word “designer”.
If we called it “applied cognitive and behavioral psychology” we would get closer to the results we want and perhaps a smidge more respect for its necessity in the age of slop UI.
Yup, I have a psychology degree and we had courses in HCI. I ended up skipping them because the lecturer (who was great, and his notes were phenomenal) was just telling project management war stories, and I'd spent the few years previous in construction listening to the same stories in a different domain.
Now, 20 years later, I'm the one telling those stories.
For Apple, and specific to Apple, when they fired Scott Forstall, there is no one to protect the software UI guys. The 2012 reshuffle in the name of "collaboration" by Tim Cook saw Design team taking over everything design, including Software. 2013 saw iOS 7 and then they spend 3 years to walk back a lot of iOS 7 UI changes. In 2016 many of their old UI leads "retired". Having no old guards to defend their turf, giving Alan Dye more rooms for power grab.
Since then, if not earlier, despite the whole UI becoming more android and cluster mess, there have been little to zero major UI changes other than new skin.
And like Apple's software, that is 10 years of resume driven development where they just want to ship features after features.
When I was getting my design degree, a large majority of what you learned was how to give and take a critique. It was central to everything you did. You were also taught things like "It's not your idea, it's an idea" that allowed you to critically think about solutions to problems without falling in love with your own designs.
This was fostered at the companies I worked for in the late 90's/early 2000's. We would have intense debates about designs, play devil's advocate (sometimes against our own designs!) in order to leave the room with the best ideas and solutions to problems. It was collaborative, energetic, peer reviewed, and we produced great designs for users.
This is not taught anymore. Critiques aren't given in school so as not to offend anyone. Much of UX Design has also transitioned to boot camps vs college design degrees which focus on pumping out students who can manipulate Figma, not students who understand the process and techniques of real design.
As such, professionally it's very hard to have real debates about design without someone getting offended or fighting for solutions that are sub par because it's "their" idea they can't let go of. The critical thinking and detail that used to go into design has been replaced with vibes.
And don't get me started on the fact that many students coming into the workforce simply don't have the design chops in the first place.
Art vs. minimalism in design
Icons are today's form of hieroglyphs for communication?
Home Trending Hacker News Hits Archive Stats Tags Links
https://blog.jim-nielsen.com/2025/icons-in-menus/
A related post linked to by the previous post:
https://tonsky.me/blog/tahoe-icons/
They needed to make room for Siri.
That was of course not at all what made the MacOS so usable. The real principles were explained in detail in Apples UI guidelines. But hardly anyone read those, so the icons have been with us for 30 years.
But at the end of the day, it was like Liquid Glass itself - looks sick in presentations, but in real life, it looks pretty awful.
Fifteen years later, I am still waiting for them to undo the damage that was done.
And when people cried what 'oh noes i cant find a needed program in that awful fullscreen start menu uwu' I just show my Vista Start Menu list which took 4 FullHD screen heights on the small font settings.
And as people in the nearby comments say the problem here is what there is a lot of newcomers who just do shit - by mere lack of knowledge or worse by being KPI driven. Just like the last update of Mozilla Firefox which added ellipsised tabs. Like... for what? On the desktop?
The problem wasn't Metro itself -- it was the ham-fisted, halfhearted integration with the rest of Windows that proved irksome. Windows 8 was the version that introduced Settings alongside Control Panel -- 15 years later, the merge is almost complete.
https://www.goodreads.com/book/show/344729.Designing_Visual_...
and the high water mark of my graphical UI experience was when I used an NCR-3125 running Go Corp.'s PenPoint as a portable/note-taking/sketching device and a NeXT Cube w/ a Wacom ArtZ as my desktop.
One of the worst trends in icons were this insatiable desire to put different icons into uniform shapes, for example, this trend resulted in tragic replacement of beautiful varied and distinct Android icons that were introduced in Android 5 to the uglicity we have now.
On iOS, beautiful SFSymbol icons that floated beautifully on flat surfaces since iOS 7.0 were encased in absolutely similar circular 'liquid Glass' blobs. Sad.
A company 1/100th the size used to design beautiful icons that clearly represented what they meant, and that company was Apple. Now I get SVG slop. SF Symbols can bite me.
But hey, tastes differ.
Things are different today though, considering how many pieces of software an average person used in 2001 vs how many today. It's no longer just word and excel, updated once in a few years, so you can learn stuff by heart and not even care what icons they put on what
I think I stumbled upon it while reading that era's GNOME HIG, that had some very similar thoughts about icons..
Apple knew this was bad 20 years ago. They just don't bother to read their own documentation:
Apple Human Interface Guidelines (2008) [1] p. 170:
> If you do include icons in your menus, include them only for menu items for which they add significant value; don’t include them for every menu item. A menu that includes too many icons (or poorly designed ones) can appear cluttered and be hard to read.
[1] https://acko.net/files/zoomer/apple-hig-2008.pdf
Though the cog for settings is, like the three-dots commonly used to represent a menu itself, is only recognised as such because it has often been used as such. The first time(s) it was used it would not have been obvious to people. So the rule isn't that a good icon should be obvious. A good icon is either immediately obvious or consistently used.
The common save icon is a good example of this in both cases: to those of us who have been around a while it is obvious because it looks like a floppy disk, to those who never used floppy disks it is obvious because that is what save icons (almost) always are.
This isn't really a case of someone overlooking previous guidance, but a symptom of an org where platform consistency is being prioritized over tradition. This means changes now cater to the lowest common denominator (iOS). Lots of recent updates this, like the new app icons, the macOS 26 implementation of liquid glass, etc. The priority is making macOS easier to adopt for people who's first devices were iPads.
It's unfortunate. Although what I find strange are some of the very obtuse patterns they refuse to change. It took them forever to add window snapping to macOS and Finder still doesn't have a cut command.
"Hey our metrics show that menu items with icons are interacted with more often."
"Oh we should put icons on every menu item and get our usage numbers up."
"I agree, bigger numbers are more impressive and will get us promotions or bonuses."
English is not my native language, so i was having a very hard time to learn how to use computer, with icon, my 8 Years old brain can kind of infer what is its meaning.