Trillions of dollars are managed in Microsoft Excel which is probably the most end user empowering software ever released in the history of computing.
Calling Microsoft software malware is so over the top that it turns off normal users and is only appreciated by people who are already committed to free software.
Malware is, by definition, computing that is done without its user's informed consent. There are many examples of malware that a laymen wouldn't readily classify as such. This page has a long list of examples where Microsoft's software or behavior is unethical. However, Excel is not in this list, as far as I can see. So you're right on that account.
A software's author can have the best of intents, it would still be malware if it's perceived to be malicious by it's users (see: state trojans, opt-out telemetry)
Requiring "user's informed consent" would make 99% of software malware for 99% of users. The majority of what software does is magic with hidden layers and data polling in the background from various sources, but that doesn't make it malware just because the user doesn't understand.
I did not. "Malicious action" is the result of violating informed consent, and as such redundant. I'm not surprised that McAfee left out "informed", btw.
>Malware is, by definition, computing that is done without its user's informed consent.
Do you manually inspect every CPU clock cycle spent, or every data connection or packet sent by every piece of SW running on your system, to make sure it was all done with your consent?
I do like to run tcpdump regularly. For a while I ran my home laptop thru an allow-listed proxy, and I was quite surprised at the results, and how little I had to allow to get what I needed done. A lot of software is phoning home. I would have thought that open source folks didn't do that, but standards or expectations have shifted. A browser that took over your DNS would have been anathema in the 90s, but now we trust the local networks even less.
And it only took abusing their position to get it there! We had spreadsheets before Excel, we would have had spreadsheets without Excel. We might have had the dominant spreadsheet software on Linux and FreeBSD if not for Excel. Maybe we would have had a couple more years of Lotus 1-2-3 being clunky on Windows but frankly that seems like a fair trade.
Excel, like the entirety of the titles in the Office Suite, was originally developed on the Mac - where it was hardly used since the Mac wasn't targeted to business at the time. Excel was ported to Windows 2.0, but like the Mac, hardly any business was using that. In the 1980's business ran on Lotus 123, not Excel.
All that changed after the release of Windows 3.0, but by that time Microsoft was already "The Dominator" as you put it. Excel 3.0 was released for Windows, Lotus took way too long to come up with a Windows release and Excel, with its GUI, had come to dominate the market. The rest is history, as they say.
I believe there was some skullduggery where M$ encouraged Wordperfect and Lotus, who dominated their markets, to focus on OS/2 as the future, while preparing to release or further promote Excel and Word for Windows with the release of Windows 3.0. (and possibly sabotaging the OS/2 effort). They also started the practice of bundling them up as "Office" which made Powerpoint a freebie, thus killing the market for presentation programs that had had competition at the time. I worked in an Egghead Software store while all this was going on.
I do know Microsoft lied and led everybody to believe they were completely on-board with OS/2 and were touting their codeveloping relationship with IBM. It was totally believable since Microsoft and IBM were joined at the hip with the PC and DOS.
Later we found out that while Microsoft was saying one thing in industry press releases and conferences, they were preparing Windows 3.0. I've never trusted that company since, and they've never failed to meet my low expectations of them.
Thing is, I remember a very different Microsoft. They used to be a great company. Then they made a deal with the devil in the 1987 timeframe and they've been evil ever since.
As with a good chunk of MS's desktop dominance, a big factor was Lotus' inability or unwillingness to embrace the common interface in Windows (what we now think of as standard menus for File/Edit/etc).
Lotus had its own proprietary menu structure generally invoked with slash. ALL programs did when Lotus was introduced. Its partisans made it hard for Lotus to abandon it, but its failure to shift into the new reality meant the oodles of new people flocking to computing as Windows exploded went to Excel by default, because 1-2-3 was impenetrable.
Word vs. WordPerfect played out in much the same way.
In both cases the MSFT offering was, in many ways, superior, but a key piece of it was the legacy player's refusal to adjust to the common menus Windows users expected.
I do not believe any bundling or OEM deals were material factors in Office's success beyond the fact that exactly those deals helped establish and preserve WINDOWS' hegemony.
"208.3. Microsoft's unresponsiveness and its actions to delay the Windows 95 license negotiations with IBM followed soon after IBM announced that it was seeking to acquire Lotus Development Corp., a direct competitor of Microsoft in messaging and office suite software, among other products.
IBM announced on June 5, 1995 a hostile takeover attempt of Lotus. On June 11, IBM and Lotus reached an agreement for IBM to acquire Lotus. IBM completed the acquisition on approximately July 5, 1995. Norris, 6/7/99am, at 35:9 - 36:12.
Among other products, Lotus offered Lotus Notes, an e-mail product, and Lotus SmartSuite, an office productivity suite, that competed with Microsoft's software. Norris, 6/7/99am, at 35:21 - 36:5."
"Microsoft repeatedly penalized IBM for competing against Microsoft
Microsoft withheld a Windows 95 license from IBM until 15 minutes before the product's launch because of IBM's preloading of competing products
Microsoft conditioned access to critical marketing support, and other terms and conditions for Windows provided to other OEMs, on IBM's not preloading competing products with the PCs it shipped
Microsoft sought to condition substantial MDA price reductions on IBM's ceasing to support competing products
Other than Office, sure? (Office is has long been one of their main cash cows. Is part of why they wanted to keep it on Mac, which is pointedly not Windows). It's also irrelevant. The point was that they used the pricing of Windows (which the OEMs) and withholding information about Windows to force OEMs to bundle Office/Works instead of competitors.
That they did it is clear. Whether they did it for simple profit (Office/Works) or to further enhance the moat around Windows (Java, QuickTime). They did it using Windows as both carrot and stick, to gain compliance. Because OEMs had no other choice.
There was also the plethora of undocumented Windows API's the Office programs were using that weren't available to anybody else. That was one of the thrusts of the antitrust lawsuit leveled against Microsoft. The best thing coming out of that was Microsoft's $150 million investment into Apple in an attempt to convince the government that they had competition and weren't a monopoly. Apple had 2% of the marketshare at the time - some competitor!
It successful by same reasons as well known flavoured sugar water, child friendly high calories processed fast food chain and tabloids. They sold lies.
That is not true, Excel is a great platform for smart non-programming type people to do interesting stuff with numbers. I often find that instead of building a reporting function into some new system, just generating a spreadsheet generator that people can take and mess around with is better for them and less work for me. There's some issues that the stats people don't like, and like all compatibility-preserving 30 year old software, there are ugly bits, but it's got real value to it. I mean maybe use OpenOffice or Google docs or whatever instead if you find Oracle or Google better than Microsoft, but spread sheets are valuable to users.
Excel is genuinely excellent. I was as big a MSFT-hater in the 90s as you are likely to be able to find, but credit where credit is due: Excel provides a level of data analysis power to a lay user that is just not available elsewhere.
Sure, tech people (cough HN readers cough) often want to trot out arguments about how "well, that sort of stuff just shouldn't be done in Excel! They should learn to use R or SQL or some other $MoreCorrectTool". That's poppycock. Excel is already on the analyst's desk. It's immensely powerful while ALSO being very accessible. Nothing else comes close.
No. And frankly, only people lying here were those who claimed open office spreadsheet is equal about a day it came out despite it being significantly worst.
There are a lot of things which are shitty about Excel but for the user it's a great program and how much better it is than Google Sheets is a great demonstration.
I think building a good spreadsheet program that is both accessible and powerful is very difficult. Excel might be the most non-trivially used "programming language" in the world.
Religious groups like GNU use outrageous evangelism like this to build their faith. Spreading absurd messages causes outsiders to shun members of the group, so that they can only find social/emotional support within the group.
And while Lotus may have declined in part due to poor strategic decisions, Microsoft is known to have engaged in anti-competitive practices against WordPerfect in the same period.
Kind of tangential but the article talks about VisiCalc, Lotus 1-2-3, and Excel but somehow fails to mention Microsoft's Multiplan (https://en.wikipedia.org/wiki/Multiplan) which was released the year before Lotus 1-2-3 came out. Granted, it wasn't that popular or successful, but it shows that, contrary to what some seem to think, Microsoft was contending in the spreadsheet arena long, long before Excel was even a glimmer on the horizon.
"Your Honor, on the way to murder the victims, my client passed by several people who he allowed to live, therefore my client is not guilty of murder."
Is there anybody who's actually been convinced to switch to the 100% pure and ethical GNU/RMS lifestyle by these articles or is it just preaching to the choir?
What convinced me was their jobs are all unionized. I would love to be the director of communications for 50K a year in Boston, but hey, at least we're in a Union
There's no denying the FSF has done good things, but I think more often than not they shoot themselves in the foot with their "wanting fair compensation for your work is evil" preaching. A market composed of honest users supporting honest developers, developing and selling software that focuses exclusively on the needs of the customer, would be the best possible system.
Actually, we encourage people who redistribute free software to charge as much as they wish or can. If a license does not permit users to make copies and sell them, it is a nonfree license.
This is moronic pseudo-religious doublespeak. They're saying it's acceptable to charge a fee for the one part of software creation that's effectively cost-free.
Few people, in practice, have been able make a living selling pure GPLv3 software. Indeed it’s not strictly impossible but it’s extremely difficult under normal conditions.
Even most of the vendors of free software today (e.g. Confluent) are using some kind of SaaS or open core model.
to sell something that is infinitely, freely replicated, means you must also add some value to the object being sold.
Unlike physical objects, where the sale accounts for the cost of production of material etc, what value does someone trying to sell a GPLv3 software add? I argue that the added value is zero, and hence, the price must be zero as well.
I talked to RMS once, and he said "Proprietary software subjugates users."
At the time, I thought it seemed a little over-the-top.
Yet since then, I have personally lost control of many of the computing devices I use. The phone was the biggest one, and I can't (don't?) avoid phones altogether like RMS. His words were actually accurate.
Now, I do think that some people try to get a message across, and are not heard, so they use stronger words until they get a response. Those messages do become over-the-top.
But I have come to realize that RMS himself carefully chooses his words.
> Yet since then, I have personally lost control of many of the computing devices I use.
No, you haven't. Or rather, you didn't lose control because of any software license. We lost control because of cryptography and enforced signature verification. Just because software is licensed as open source doesn't mean it allows for third-party firmware modification.
Maybe it's because I came up through the RE scene, but the false association between open source and ownership or security has always stood out to me. Many of the modding and hacking communities that orbit proprietary software are/were more vibrant than anything that exists in the totally open source space.
> Or rather, you didn't lose control because of any software license. We lost control because of cryptography and enforced signature verification.
I'm not sure what you mean by this. It's the law that's a barrier, not cryptography. If you break the DRM on a game you can still face legal consequences, even if the company that licenses it doesn't support it anymore.
Copyright laws are not what is stopping custom firmware projects on many devices. It’s the inability to flash non-signed firmware. Legal considerations in general have minor pragmatic relevance for modding projects that are not lazily redistributing copyrighted assets or circumventing copyright protection. If you’re engaging in piracy then you’re not just giving people more control over what they own.
Exactly, you just named a subset of all the things you could do to a machine, partitioned by copyright protection. I'd agree with your general point in the context of power users, but it's worth noting that the average Joe will try to faithfully follow company policy as well as the law.
DRM protection sometimes rears its ugly head, and that is worth objecting to, but as someone who has written unpackers and game mods, it’s rare nowadays that DRM prevents modding, unless you’re creating unethical multiplayer hacks. Code that is relevant to modders is usually contained in unprotected script and library assets.
To run afoul of the DMCA a modder needs to (1) strip a binary of its copyright protection (2) distribute the unpacked executable or the unpacker tool (3) likely facilitate mass third-party piracy (more damages = more lawsuit potential). It’s principally warez and hacking groups that meet these criteria, and they probably ought to be sued.
I’ll add a point in your favor and say that for non-games DRM is often a big problem when reversing or patching code.
Free software licenses are just an imperfect mechanism for trying to secure software freedom, which is the real thing the movement has always been after. And the GP's point that the loss of software freedom, even in the abstract, has led to more and more losses of control over particular behaviors on our devices. The technical means by which this encroachment is enacted are insignificant. The substance of the issue is political.
Yup, GPLv3 specifically requires manufacturers to provide any signing keys needed to get code to run:
> “Installation Information” for a User Product means any methods, procedures, authorization keys, or other information required to install and execute modified versions of a covered work in that User Product from a modified version of its Corresponding Source. The information must suffice to ensure that the continued functioning of the modified object code is in no case prevented or interfered with solely because modification has been made.
> If you convey an object code work under this section in, or with, or specifically for use in, a User Product, and the conveying occurs as part of a transaction in which the right of possession and use of the User Product is transferred to the recipient in perpetuity or for a fixed term (regardless of how the transaction is characterized), the Corresponding Source conveyed under this section must be accompanied by the Installation Information. But this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product (for example, the work has been installed in ROM).
I think this is a good argument for the advocacy of GPL-3+ prevalence. It’s also legal, I think, to publish GPL-3 firmware source code, and to disallow custom firmware flashing, as long as the copyright is self-owned. (IIRC the provisions of the GPL don’t give legal standing for suing the author over self-violations.) In this case, would you call the software or code proprietary (which many confusingly do) or would you call the device proprietary?
Sometimes, yes, but not intrinsically. AOSP performs OS and app update authenticity verification via predistributed public keys, as do package managers. In theory, a phone’s firmware can be completely open source and still be unmodifiable. Proprietary software isn’t any more opaque or protected than open source software.
Sorry. I meant to write that closed source software isn’t opaque, technologically. (More opaque, yes, opaque... no.) Legally, it’s different. But closed source compilation alone doesn’t grant a program special privileges or confidentiality or non-reversibility, which is sometimes assumed. It’s just machine code or bytecode either way.
I prefer OSS in every case where it is viable but there are just no alternatives. Not because RMS, just when shit inevitably breaks I have all the parts in the view, not black box making smoke. Or even occasionally just fix it myself and throw a patch.
I've wondered that myself. In my experience, people were practically born into the GNU lifestyle or they weren't. I've never known someone to convert past the age of 20.
I switched to GNU/Linux after reading essays from the "Free Software, Free Society" book. It took years, but I did it. Also, I'm using an FSF-endorsed OS on my phone right now (Librem 5 with PureOS).
I feel like really outre levels of evangelism, including the linked article but also & especially including religious evangelists who go out of their way to try and preach in inhospitable locations (e.g., preachers on left-leaning college campuses yelling about gay people) are partly or even MOSTLY in it FOR the backlash.
Partially, yes, switched to Linux. I always thought RMS being a fun and entertainig but ultimately a weirdo. Turns out he was right all alo g and I was an idiot.
I'm all for calling out large corporations and bad actors on their misconduct, but this article seems bigoted against Microsoft and their products, and hypocritical at times.
For example, (referencing the "Jails" section here), the author says, "Jails are systems that impose censorship on application programs." Not sure if I'm interpreting this next quote from the GNU.org site (https://www.gnu.org/distros/free-distros.html) correctly, but it seems like this is exactly their definition of a "Free GNU/Linux distribution" is meant to do: "They will reject nonfree applications, nonfree programming platforms, nonfree drivers, nonfree firmware "blobs," nonfree games, and any other nonfree software, as well as nonfree manuals or documentation."
"They will reject" seems an awful like "won't run on my distribution", and not because it's not compatible because of runtime or kernel differences, but because of its 'being nonfree'.
From Merriam & Webster:
BIGOT: a person who is obstinately or intolerantly devoted to his or her own opinions and prejudices.
I love free software. I don't have to pay for it, I can share, customize, and use it as I want without the author of the software ever needing to know or care, because he did it out of either love for software or because he had a need to fulfill and was nice enough to release it to the world.
I also use paid software and software containing "nonfree" software blobs. I have an Android because it works for me. I use Spotify on all my devices, and pay for Premium, because I find it convenient, the same reason I use GNOME and VS Code and drive a used Chevy.
I value my time and quality of life too much to be extreme "beCaUSE iT wASN't FreEEE".
Just my 2 cents. I'll have that with a side of Kale please.
I run Windows 11 on my personal laptop. Best part is it's a Framework Laptop, it's 100% compatible with Linux and has official support for Ubuntu and Fedora, and yet I choose to run Windows 11 on it. I prefer the Windows UI to the abomination that is GNOME and I'm also able to do basic things like sleep and hibernate my machine without it freaking out.
I prefer the gnome ui to the abomination that is windows :) I'm also able to do basic things like sleep and hibernate my Ubuntu machine without it freaking out.
I actually like Plasma a lot as well out of the Linux desktop environments, it feels like the only one that actually cares about design and UX. It's a shame that GNOME is the 'default' Linux DE and often the only one with real support from distros.
That's one of the things that I really can't understand. Gnome is awful and the people who work on it are so out of touch with what users actually want.
Plasma understands that; as well as PLENTY OF OTHER DEs but your casual linux observer just thinks "wow gnome looks ugly" and it does, it always has, i still have absolutely no idea why anyone would roll a distro out with that... mess.
I imagine it was this one [0]. I was confused by it at first because it amounts to more than 100%, then I understood it's a multiple choice question, so the correct affirmation would be around 50% of spftware developers use Windows in a professional capacity, but that doesn't exclude them also using Linux, macOS or BSD.
any developer developing for windows is on windows
most game developers are on windows
whole tech sectors have been captured successful by Microsoft in the past and most developers there have to use Windows weather they like it or not, some of them are _really_ big (but also not super vocal or visible on GitHub as they tend to be quite proprietary in all of their actions).
etc. etc.
Idk. how realistic 75% is (e.g. what it counts as developer who/how people where asked etc.) but it's probably a high rate anyway.
EDIT: Also also probably using one doesn't exclude the other and doesn't necessary exclude them running it in a VM ;=)
When you say ‘Word’, do you mean the app, the online one, or the Teams thing? They all behave differently and aren’t fully compatible. It’s very frustrating how formatting changes between the Words.
I’m bound by a regulator to format a certain way and it’s brutal when using a full MS stack.
I really do hate is MSFT's naming conventions AND their propensity for calling two different products by the same name for marketing reasons.
I mean Microsoft Word, the word processing tool that runs as a native app on one's computer. The degree to which the online or Teams versions do not play nice with the "real" one is a HUGE own-goal by MSFT and reason enough to avoid them if you can. If you're going to have an online tool called Word, it damn well needs to work EXACTLY LIKE THE REAL VERSION.
I've used both extensively and interchangeably. The fact that Microsoft got "Word" running in the web, and embedded inside another application, and made it all work mostly seamlessly is downright amazing. I've yet to encounter any issues with formatting between the three, and am constantly surprised at their ability to be consistent.
Now where it all goes pear-shaped, is not the technical side of it all, it's the UX and marketing teams that got involved. One big example is the "save" dialog that's been butchered beyond describing (even in the standalone desktop version of Word) in order to promote OneDrive and other microsoft offerings.
My issues: Pictures, headers and footers, the amount one can fit in a page (varies between versions). What will happen to the Teams version if I download it, edit in the native app and save?
Downloading a lot of files off Teams usually fails and gives a load of html files.
The ‘integration’ nag for one drive which can’t be dismissed or enabled but seems to work.
It’s an absolute mess and I’m very surprised to encounter someone who finds it works well.
> Honestly, Teams
> Exchange / Outlook, for group scheduling
Are you sure?
We are a ~15 person company and there isn't a (work) day where someone doesn't have problems with teams. I can't say for sure because I hadn't used it a lot but jitsi seemed way better and google while having far worse UX was more reliable. Like screen sharing stopping working (without any error or anything just stand bild) during a meeting is pretty common, or abstruse things like one time the chat was just missing until I restarted it or another time there was a partition of rooms until everyone left the meeting and then reentered it again.
Similar problems with outlook are also not that uncommon (but more rare), like e.g. calendars not syncing, the only things really speaking for it (compared to some other good team calendar applications) IMHO is it's integration with teams... which is worth something, I mean it's pretty convenient.
We're about the same size, and don't have issues with it really at all, but we use it only for:
* Calling each other;
* Ad-hoc meetings internally, because our normal meeting platform has become cumbersome (GoTo);
* IMs
I see it'll do other stuff, but we're not really using those aspects. And I should also be super clear that I'm not saying Teams is actually GOOD. I have just found it to be better than any of the alternatives I've seen or used.
That's also very much how I feel about Outlook/Exchange. I mean, look, I'm a Mac guy. I use the Mac mail client almost but not quite exclusively; I also have an RD session up to a Windows VM all day every day for other reasons, and run Outlook there. The gap that drives me to Outlook/Exchange is scheduling. It just works, and works in a way that nothing else I've seen approaches -- even when messaging people outside our org.
I wish Apple calendaring was this good. I really do. But it's not.
>"They will reject" seems an awful like "won't run on my distribution",
No it doesn't. You are free to install whatever software you want. None of these systems will prevent a user from using non-free software. They just won't encourage it.
I find GNU's position too dogmatic, but it's not as bad as you pretend.
I thought bigotry was scapegoating a minority group because of the actions of a few members.
Do you value the time and quality of life of the billions who can't drive a Chevy, nor will ever have a device that can run an Electron app or $10/month for Spotify?
It's not all free as in beer you know, free software is meant to serve a free society by allowing everyone to participate in the social construction of technology, not provide consumer luxuries. I think the term you're looking for is "open source"
I've been in IT since the 1970s. The decrease in women in the field is striking. The article you cite is correct to make the connection to CS degree representation. Prior to the 1980s, CS degree programs were nascent if they existed at all. Not requiring this qualifitcation allowed an amazing diversity of people to enter the field.
Steve Jobs was in a way quite typical of the sort of people who showed up in the IT field. He would have a tough time getting hired today.
but open ideals are enlightened ones, so it cannot be bigotry per your definition.
I suggest staring deeply into why culture and natural language work like they do (or used to do, seems only stubborn 'artistic' types are working with culture like that nowadays)
> I thought bigotry was scapegoating a minority group because of the actions of a few members.
That's certainly one common recent use of that term. Although more often than not it's just "I don't like them but don't really have argument that passes first smelling therefore they are bigots"
> "Jails are systems that impose censorship on application programs." Not sure if I'm interpreting this next quote from the GNU.org site (https://www.gnu.org/distros/free-distros.html) correctly, but it seems like this is exactly their definition of a "Free GNU/Linux distribution" is meant to do: "They will reject nonfree applications, nonfree programming platforms, nonfree drivers, nonfree firmware "blobs," nonfree games, and any other nonfree software, as well as nonfree manuals or documentation."
It's not; first (as referred in article) is preventing user from using the machine as they see fit.
Second one is not accepting software into distribution that tries to take user's freedom to modify it; they don't stop you installing non-free software in free distribution, they are just not shipping the non-free software. It's a stance on not promoting (by including and making installing trivial) the non-free software but you are still free to install on your own.
Going by tired
> "They will reject" seems an awful like "won't run on my distribution", and not because it's not compatible because of runtime or kernel differences, but because of its 'being nonfree'.
Nope, it explicitly means "will not distribute, with their distribution, for free". Software actively stopping you from running other software would actually be nonfree, as the whole deal with OSS is "do with sofware what you want" and GNU deal being "and pass same priviledge forward"
> "They will reject" seems an awful like "won't run on my distribution", and not because it's not compatible because of runtime or kernel differences, but because of its 'being nonfree'.
It doesn't sound like "won't run on my distribution" to me. It seems that you intentionally took this line out of context just for the argument (or to defend MS).
The full paragraph:
> These distros are ready-to-use full systems whose developers have made a commitment to follow the Free System Distribution Guidelines. This means they will include, and propose, exclusively free software. They will reject nonfree applications, nonfree programming platforms, nonfree drivers, nonfree firmware “blobs,” nonfree games, and any other nonfree software, as well as nonfree manuals or documentation.
Next paragraph:
> If one of these distros ever does include or propose anything nonfree, that must have happened by mistake...
"This means they will include...". Not "This means they will run...".
"include or propose...". Not "execute or install...".
It clearly means that they don't include nonfree stuff with their systems. Not they don't allow nonfree stuff to run.
Well, you have to pay for new android phone every couple of years, otherwise your out of date android firmware will be hacked by hundreds of RCE. You being tracked and targeted without possibility of refusal. Tour phone is not yours.
This has happened because Torvalds let hardware manufacturers to tivoize kernel. This has happened “beCaUSE it wASN't FreEEE as freedom rights but as a cheese in mouse trap"
Arguably Linux would not have grown as ubiquitous if it's license didn't have the plugin loophole. Perhaps a BSD, Minix, or another would have arisen instead.
Or it would just be a Windows monopoly ecosystem. Without that loophole perhaps no open source unix-alike product would have been able to out compete Microsoft in the server space which is what lead to it's viability as core of a phone OS.
This is why I much prefer the MIT license though still respect GPL. Copy left is cool and all but it suffers from similar problems as paid software - you have restrictions on what you can do with it. The MIT license captures the spirit of “do whatever the fuck you want”
I think the main problem with MIT is that if you offer the largest industry players the opportunity to “do whatever the fuck [they] want” then they will. Historically, this has not been great for users.
As an obvious example, I honestly can't imagine the Linux kernel having had the impact that it has had if all of the myriad corporations building on it weren't forced to contribute their work back to it under the same conditions.
On the other hand, plenty of software nowadays is permissively licensed (in fact, the overwhelming majority outside of Linux and the GNU project) and everyone from individuals to huge companies contributes to it for a variety of reasons.
It's basically the same as with the paradox of tolerance:
> The paradox of tolerance states that if a society is tolerant without limit, its ability to be tolerant is eventually seized or destroyed by the intolerant.
You're right too. Microsoft tools have enabled me to make a lot of money over the course of my career, but they're seriously neglecting quality of life features, and they're buggering Windows for developers.
On the side, I've even learned to like kale on pizza. The Kale-Zer Soze at Central Pizza in Seattle is outstanding.
> Arch has no policy against distributing nonfree software through their normal channels, and nonfree blobs are shipped with their kernel, Linux.
Debian is a no go
> Debian also maintains a repository of nonfree software. According to the project, this software is “not part of the Debian system,” but the repository is hosted on many of the project's main servers, and people can readily find these nonfree packages by browsing Debian's online package database and its wiki.
Sorry Gentoo
> Gentoo includes installation recipes for a number of nonfree programs in its primary package system.
BSD
> FreeBSD, NetBSD, and OpenBSD all include instructions for obtaining nonfree programs in their ports system. In addition, their kernels include nonfree firmware blobs.
I really appreciate GNU keeping their ground. We need them to keep the Overton window correctly framed.
Nonfree drivers are not OK.
Nonfree BIOS/EFI programs are not OK.
Nonfree kernel blobs are not OK.
We may use them out of necessity and pragmatism (some nonfree software is better than 100% nonfree software! an Arch user still has more intact freedoms than a Windows user), but we should still reject them ideologically and remember that they are tolerated but not accepted.
Specialized use or small audience or should be accommodated. I'm specifically thinking in terms of handicap accessibility. I need speech recognition in order to work effectively. The only good speech recognition environment for people with disabilities is Dragon on Windows. Everything else simply converts speech to text. I need something converts speech to tax then digs into text boxes and GUIs to enable editing and recognition corrections by speech.
The same also applies to copilot. I've trained myself to give it prompts to generate code that matches what I would've generated by hand years ago. But without copilot, I wouldn't program. Simply not practical.
You’re replying to someone named “fsflover”. This is a bit like trying to convince someone named “maolover” that capitalism has some advantages.
The idea that non-free software is unethical and should be abolished is the bedrock of all of the FSF’s philosophy, and is held almost as a religious view for RMS and his adherents.
Not sure what their nick is supposed to mean. And I'm more interested in understanding why they feel so strongly, and whether their view extends elsewhere, than trying to change their mind.
Religion is completely unnecessary here. It can be purely practical. By relying on nonfree software you loose your possibility to escape walled gardens (Apple), could be forced to pay more for support, or can't get any support at all (some Android phones, old Apple phones) and so on.
It's unethical, because the developers of nonfree software have too much power over the user. They create an artificial monopoly, where nobody except them can provide support in case something goes wrong. Thanks to this monopoly, for example, you have to buy a new Android phone every two years.
All these questions are answered in book "Free Software, Free Society" (with better explanations than I can give).
> mechanical machine that was still covered by patents or trade secrets
Debian does not clearly separate free and nonfree repositories. Users might be mislead into using nonfree packages without their knowledge/consent. I once was misled like this.
Today, I am using Debian myself and avoid nonfree packages knowingly. However I am glad that FSF keeps reminding users about it. Note that they do not prohibit the usage of Debian. They just don't endorse it.
What if the bios, drivers, and blobs were source-available? Obviously a true free license is preferable but I can see why commercial vendors are hesitant to allow anybody to make whatever changes they want. From what I understand the GNU project would still reject that on ideological grounds even if it addressed a lot of the theoretical problems (running code where you have no way of knowing what it’s doing)
>commercial vendors are hesitant to allow anybody to make whatever changes they want
Why? If a vendor sells a fingerprint scanner or something why do they need the driver to be closed source? Why does somebody tweaking the source bother them?
I've never really understood this. Sure, Microsoft word needs to be closed source but why a driver?
I always figured a lot of it is covering over garbage hardware in software. Whatever "clever" feature it had doesn't actually work and they fake it in the drivers, or they're having to patch around a maze of different buggy hardware revisions. Nobody wants to show the world how that sausage is made.
I think a lot of it is also just a general amount of fear for (from the companys perspective) no benefit. There's a lot of fear that if the code is visible, you could be sued for its contents or it could be otherwise used against you in court, competitors could "steal" your effort, etc.
Plus there's just a lot of belief in the nonfree circles that the software freedom movement is just religious zealot hippies who think software should never be paid for and that microsoft programmers should work for free and starve, and that anyone who does develop novel software should have it immediately ripped off and copied by competitors and make no money because capitalism is bad.
Then of course that's better! But still not good enough.
If I fix a bug or desire some new functionality and implement it for myself, am I free if I cannot legally share it with my neighbors? Certainly not.
Also depending on exactly how they release the source, you may still not know that's what the blob is actually doing. I suspect many vendors, even if forced somehow to provide source, would try to not allow (or make very annoying) building it or signing it. And if you can't build it yourself, or if you can but can't run it and the build isn't bit-perfect reproducible, then you can't know the blob is actually the output of building the source.
You could decompile but let's be real, if security researchers looked at Heartbleed not once but twice, specifically in the context of it being a bug, and both times asking if it's a security vulnerability, and said no - then I am not going to believe that a human inspecting decompiled machine code is going to be very good at finding well-hidden malware in blobs.
A pretty sobering list. Some of the language is charged according to some folks, but saw no errors per se. None of it is news if you've been paying attention for a few years.
I no longer need to worry about Windows, but now wondering what to do about UEFI. I've heard about things like coreboot, but doesn't seem to be offered by mainstream manufacturers. Not sure I wanna risk bricking my machine. Has there been any recent progress here?
I think it's difficult on Intel to be fully free because of Intel ME. As far as I know, it's not possible to fully be rid of it, so you'll always have some proprietary blob running a possible backdoor.
If you want to be fully free, you may be interested in the Raptor Talos II, which is based on OpenPOWER and has free firmware. It is pretty pricey however.
I don't think we talk enough about how U.S. taxpayers are footing the bill for Microsoft's technical debt.
It's shocking to spend time with Federal Workers and see how much of their days are wasted waiting for anti-virus and other bad closed source software, and then to realize we all are paying taxes to support this! Same goes for Oracle. BG and LarryE got the country to fund their yachts and stuck us with the bill.
I'm gonna run for Congress under a platform of banning any Federal money from being spent on closed source, non public domain software.
And that's just a fraction of the waste. The government spends millions just managing compliance with software licenses. (source - extrapolating from my experience at a 3 person Microsoft shop 15 years ago) But this isn't technical debit - this is by design.
Even if one believes that zero Microsoft software is malware, it seems indisputable that Microsoft software is the single greatest attractant for desktop/laptop malware, from any source. (Perhaps Android is an even greater malware magnet, but not on desktops and laptops.^1) The goal is popularity and profit, not user control (cf. non-commercial options).
The ideal user to Microsoft appears to be an uninformed one, not interested in control and apprently willing to reliquinsh it. Not all Microsoft users are uninformed or disinterested in control. There are profound exceptions. Some of those exceptions will read this comment and respond accordingly. But hundreds of millions of non-exceptions will remain uninformed and disinterested, to the benefit and delight of Redmond.
Obviously, malware can thrive under the auspices of uninformed, oblivious users, Microsoft's ideal customers. Microsoft will purportedly protect them, when it wants to and assuming it can; of course, it won't protect them from Microsoft itself. For those Microsoft employees and users not reading, commenting and/or voting on HN, enjoy the ads injected into Windows. Microsoft is happy with the result.
(The point that commenters/voters appear to be missing is that it is the nature of the customer/user that creates the attractant for malware as much if not more than the software itself or the size of the user base. GNU regardless of what one thinks of them is AFAICT advocating for user control in this communication. That implies users who care about control, not the average Microsoft user. Users who are paying attention and who do not blindly trust companies like Microsoft. One can argue it is the "insecurity" of the software or the number of users that creates the attractant, and there is certainly some truth there, but it makes no sense if, e.g., the user base is a large number of paranoid users who read HN. They are going to avoid/prevent/detect malware better than a random sample of Microsoft users. User behaviour is relevant. If 9 out of 10 users who run program X on operating system Y are careless, then program X and/or operating system Y is arguably an excellent target for malware.)
Most people use Android nowadays and a lot of the older generation is seeing a lot of Mac usage as well as iOS.
If MS dealt with all their technical debt and built a brand new OS following proper disclosure as well as OSing most of their kernel, you'd see that it would be much harder to exploit.
Think of linux that runs over 90% of all servers in the world. How many ransomware attacks have they been hit with?
This is just a list of compromises where Microsoft is prioritizing something other than what this author wants. It turns out a lot of users want to have a secure system, not lose all of their files when they forget a password, can work with legacy systems, etc.
Calling software malicous because it prioritized security over freedom just reads to me as being ignorant.
No offense, but has this guy used Google? Because it's much worse. Microsoft is not innocent, but Google makes them look like child's play. There are 3 tech companies (in reality) that are juggernauts. Apple, Microsoft, and Google. Apple at least pretends to care about privacy (I don't believe them) Microsoft somewhat pretends (I don't believe them) and Google doesn't even pretend (I actually believe them.) Pick your poison.
Looked at the Apple one. Parts of it seem really childish.
I can figure out iPhony is an iPhone, but I'm not certain what iBad and iMonster are supposed to reference. I could dig into the source references and try to figure out WTF they're trying to convey, but that shouldn't be necessary.
Regardless of context, it harms one's argument (and is blatantly disrespectful) to refer to people or groups or items with labels other than their actual names. That kind of foolishness has the same level of rhetorical value as MAGA types calling the opposition party the "Dummycrats" or whatever.
>Nonfree software is controlled by its developers, which puts them in a position of power over the users...
The implication is that FOSS developers have no control over their users. Sure, you COULD fork your favorite tool's repository and spend many hours trying to figure out how to make it do 'what you want instead of what the developers wanted'; but very few people are really willing and/or able to do that.
Nearly all software (open source or proprietary) except trivial programs, are controlled by the developers who build it and maintain it. They are given that power by the mere fact that most programs are sufficiently complex that few others can and will assume control instead. If someone does (and that someone isn't you) then the new boss is same as the old boss (to quote a line from a song by The Who).
Sure, you COULD fork your favorite tool's repository ... to make it do 'what you want ...’ few people are really willing and/or able to do that.
Free software makes it possible for me (my business, etc) to pay someone other than the original developer to make the software do what I want. No proprietary license gets me that.
In the context of personal use software by an individual it may not make sense to pay, but in the context of groups of individuals or a business it becomes viable.
While it'd take many hours to figure out how to make some random FOSS code to do what I want if what I want is anything particularly interesting, finding out where to comment out the code that eg phones home, or makes annoying noises, and then sharing that patch with everybody else is much easier and more realistic. Sure, it doesn't entirely remove the power of the developers and particularly the people who control the official releases page, but it feels like a real, meaningful check on their power. It becomes even more meaningful if you're not just some rando but maybe the "user" in question is an org who can actually devote resources to maintain changes to some FOSS code, saving you from having to enter into some sort of partnership with the developer.
Are you saying it makes it easy to maintain patches parallel to the existing software? The difficulty keeping forks up to date (losing maintenance benefits of upstream work) and "re-basing"(?) the custom changes in a clear and clean way is my biggest stumbling block to "just fork it".
The actual entities that control people through software are very rarely developers, but institutions and corporations. And one of the things that software freedom really does allow is for those developers to jump ship without abandoning the software when those entities become abusive or change the deal.
And this does happen! The response to Oracle's anti-F/OSS, user-hostile moves when they bought Sun is a powerful example.
Free software also means that you can pay developers to take the software in a different direction if you can't do so yourself.
There are lots of user-serving forks in the world. Cinnamon and MATE really did serve Gnome 2 users who were pissed off and disappointed about what they saw in GNOME 3. LEDE really did resurrect OpenWrt by giving people a way to push the project forward when development had stagnated. yt-dlp saved a bunch of youtube-dl users who would have been left in the lurch.
Of course creating and maintaining a fork requires non-trivial resources. But forks do happen and their possibility absolutely makes a difference to end users, lots of the time.
I didn't mean to imply that there are not numerous examples of open source projects that have been successfully forked to provide meaningful value in a whole new direction. However, the article seemed to be saying that open source magically takes control away from the developers and gives it to the end users. I just think that is the exception instead of the norm.
Hmm, I may have misread you. I'm still glad to have emphasized the points I did, because I think they're important for the broader discussion. Sorry for any misunderstanding!
> The implication is that FOSS developers have no control over their users. Sure, you COULD fork your favorite tool's repository and spend many hours trying to figure out how to make it do 'what you want instead of what the developers wanted'; but very few people are really willing and/or able to do that.
Few people are willing to do it because the kinds of restrictions/annoyances imposed by free and open source software are much less severe than that of nonfree software. Sure, there might be the odd feature or two missing in a FOSS project that you would like but that the author blocks from being added to the project for XYZ reasons. I'll concede that those kind of decisions give the maintainer of a FOSS project some semblence of "power". But it's not like proprietary software, where trackers, atrocious adverts and news links (looking at you, Microsoft start menu) and other arbitrary restrictions are a continuous reality. If those things were in a FOSS project, you can bet that _someone_ would make a fork with that stuff removed.
Plenty of companies including every one I’ve worked for maintain their own forks of open-source software, often while waiting for maintainers to apply patches.
I’ve also contributed patches upstream to open source software plenty of times and the majority of time the maintainers have been willing to accept it.
Tons of Linux distros maintain their own set of patches on top of loads of software to make it work better with the rest of their system.
Sure, there are examples where it’s infeasible to change things but in practice it’s a lot more likely than with proprietary software.
On one hand, I've been one of Microsoft's biggest fans. I've been using Windows since it first came out, then NT, then 95, then 2000, and so on. I've been using Visual Studio since 2003. They used to be my favorite products.
And yet I hate what they're doing now. I turned off automatic updates, or so I thought, and returned to my machine one day last week to find it had restarted anyway. I'd spent a fair amount of preparing a commit and most of my work had vanished. Another time I returned to an unwelcome restart and I had actually lost a fair amount of work. Sure, I retyped it, but really?
On the same machine, I have another account on the same machine that I use for testing. Every single time I sign in to that account, I have to start Task Manager as an administrator and then kill Windows Compatibility Telemetry, even though I have allegedly(?) disabled it several times. Otherwise it consumes so much CPU that I can't get my work done.
Also, Visual Studio is a hot mess. Why do I have to wait when I drag and drop an int into the watch window? Why does my machine slow to a crawl when I'm running debug builds? Why are there so many bugs with breakpoints? A lot of the time, I have to use page up/page down to see the breakpoint appear even though I hit F9. Why is it agonizingly slow to type things when I'm debugging? Why do I fantasize about finding a new job that doesn't require me to use Visual Studio? Visual Studio has become my least favorite piece of software. There are days when I just write code in Notepad++ and only use Visual Studio to build.
It's frustrating. The seeming inability to exert developer-level control over Windows makes me feel like my devices aren't mine. For fuck's sake, I had to tell Windows that my internet connection was metered in order to get them to stop with the updates.
Yes, the article might be hyperbolic, but the whole jail argument is starting to line up pretty well with reality. Dear Microsoft, please stop fucking with everything.
I've worked on a lot of large and complex projects (mostly kernels) using just vim, git, gdb, cscope, and ctags for development for the past 15 years. I think I've encountered maybe one or two quality of life vim bugs in that time, and a few annoying gdb bugs.
Like AI, until someone shows me something that can come up with novel solutions to deeply technical problems, simple tools and intelligence on the part of the programmer is what works.
Unfortunately I don't have any choice in the matter of compiler or platforms. I do some work work on Linux and I use GCC there. It feels much more sane.
Visual Studio makes me want to take more blood pressure medication.
This is tier one thinking. Maybe top taking the blood pressure meds and get a little bit more angry and your mindset will change.
Until then, enjoy your MSFT environment that they've created for the lowest common denominator. You try the code-gen / no-code tools yet? Maybe they'll make you happier. Developers, developers, developers!!
you still can invoke msbuild manually. If I really need to test on Windows, most of the time it's a CMake project so I just generate build files for VSCode, go back to whatever editor I like and `cmake --build .` to launch the build job.
The forced updates are truly outrageous. It particularly irks me since I always chose them over apple for respecting that after a purchase transaction ends, the product is mine, and I need no longer deal with them if I don't care to. And bending over backwards to maintain backwards compatibility. It was an enterprise focus. They have gotten some bad press for the auto updates, so I'm sure they have made a calculated decision to stick with them. And it's hard to blame them looking at how much money they've raked in since this pivot early/mid 2010s.
For what it's worth, it's (still, as of writing) possible to turn off forced autoupdates via Group Policy if you buy and use Windows 10/11 Professional.
Any power user would be well served to reuse one of their old 7/8 Pro licenses or buy a 10/11 Pro license instead of cheaping out on a Home license.
The real problem with "it's still possible to turn off forced autoupdates" is that you don't know whether you can turn it off, due to Windows ignoring explicit "turn X off" switches. Why should the user trust you to respect any other switch?
Also, the user should be able to assume that there aren't multiple switches for the same action. As a general rule - I know this is Windows.
If you want to like Windows again try the last verison of Windows 10 LTSC. Its pretty damn amazing. All of the crapware is stripped out and yet it still has the insane backwards compatiblity and ability to be installed on a huge range of devices (very underrated features of Windows). The battery life after installing on my laptop is insane. It boots up insanely fast.
I have to imagine there a bunch of software engineers at Microsoft who have worked their asses off and are proud of what the core of Windows has become but are also super frustrated at all of the BS that is loaded on top in 99% of installs.....
Windows 10 LTSC - I couldn't agree more. If this version doesn't exist I'd be only using Windows for gaming at this point. I still heavily modify LTSC but it's a big improvement from the unusable HOME version.
> There are days when I just write code in Notepad++ and only use Visual Studio to build.
Judging from what I've seen from Windows game programming streamers like Casey Muratori and Jonathan Blow, it's not an uncommon practice to live in your preferred editor then flip to VS for build/debug.
It all seems very strange and ridiculous to me coming from a gnu/linux/vim/screen/make/gdb background. Obviously not so much having separate tools, but more the despite having a massive beast of an IDE purportedly providing All The Things, you're still context switching to/from your usable editor.
Those streamers also often resort to using RemedyBG for debugging instead of using VS's integrated debugging. They're basically living without an IDE, while paying all the overhead for an IDE, because the IDE is such trash.
> On one hand, I've been one of Microsoft's biggest fans.
Microsoft helped lift me my family out of poverty. I grew up in Seattle, in the shadow of Microsoft and their Redmond campus. My father was a laborer who helped built Bill Gate's mansion, and who helped built parts of their Redmond campus.
He was incredibly, incredibly, proud, that I drove to work everyday down the streets he had helped plant trees on. He told almost everyone he met that his son worked at Microsoft.
Microsoft did a lot of good in the 90s and early 2000s[0], they created a standard platform which the world built upon. They created a stable foundation for thousands upon thousands of developers to build their livelihood. They paid for generations of children to learn programming in free or next to free programs across the country (and across the world). Since their missteps and the rise of mobile, the industry hasn't had a stable development platform to build upon, and we've collectively wasted untold billions of dollars in productivity on crappy cross platform solutions.
When I was at Microsoft, the customer was king. I put a stop to plenty of bad behavior by saying "is this what is right for the customer?" and if the answer was no, code didn't ship.
But Microsoft is a fear driven company, they were afraid of the rise of the tablet so they made Windows 8, and then they became afraid of Google so they constantly attempt to go for ads. (First they went for search, but then eventually realized Google isn't a search company...) Heck the original reason Explorer was supposed to get tabs was so ads could be shown in new explorer tabs[1].
The other problem is, Microsoft got rid of Software Development Engineers in Test, and they replaced them with devs doing test. But at Microsoft, SDETs served as the last line of defense for the users, as an SDET it was made clear to me from day 1 of my employment that I had the power to stop shipment of software if I felt the code wasn't up to snuff. As a junior SDET I only ever used that power once, I was damn well scared, and when I spoke up and said we had open bugs that I personally considered ship blockers, the lead dev said no problem, the release would be delayed until his team fixed the bugs.
No one has that role anymore. Before I left MS around 2016 or so, I saw a fellow developer instructed by the design team to make a UI component behave as if it followed the MacOS design guidelines. The (junior) dev didn't realize that was what he had been told to do, and it obviously took a long time for him to beat the Windows UI controls into shape until they worked like the MacOS UI controls, and when I asked him why he was doing this, he said he was too scared to push back.
That is what SDETs were for. To hold the bloody line. Facebook's "move fast an break things" motto gave the industry an excuse to gut dedicated test teams, and now all software everywhere sucks, but no one under 30 realizes anything is wrong because from their perspective things have always been shitty and broken.
[0] Yes yes and Linux bad lawsuits etc etc, separate discussion.
[1] The ads were to be for Microsoft Store apps IIRC.
> The other problem is, Microsoft got rid of Software Development Engineers in Test, and they replaced them with devs doing test. But at Microsoft, SDETs served as the last line of defense for the users, as an SDET it was made clear to me from day 1 of my employment that I had the power to stop shipment of software if I felt the code wasn't up to snuff. As a junior SDET I only ever used that power once, I was damn well scared, and when I spoke up and said we had open bugs that I personally considered ship blockers, the lead dev said no problem, the release would be delayed until his team fixed the bugs.
I worked for almost 27 years, for a Japanese optical corporation that is synonymous with "Quality."
Their QC department was the most powerful one in the corporation, and all the devs lived in fear of failing QCT, as that would have C-Suite-level notice.
In the US, there used to be a joke, saying "If you have 'Quality' in your job title, your career is over." At this company, it meant you were just short of a God.
That said, it tended to make life stressful for us. As an American development team, we were often treated as "Bad Software Cowboys." Most folks here, would consider us to be completely uptight, anal-retentive, risk-averse, babies, but they thought of us as reckless.
> In the US, there used to be a joke, saying "If you have 'Quality' in your job title, your career is over." At this company, it meant you were just short of a God.
I would welcome this as a cultural change in the US. It just sucks living downstream of people dumping tons of cowboy code into the river.
MSFT seems to be a lost cause these days. Whenever I open up VS2022, it takes almost an entire minute to load. As I recompile things, it will fail to report errors, which means restarting the application (which seems to "fix" it).
If I open SSMS, it also takes upwards of a minute to load. Some things I used to take for granted, like being able to see the cursor when scrolling with the arrow keys, instead makes the cursor invisible, so now I have no idea where my cursor lands when I release the key. When a connection to a DB bonks out (which happens at least once a day), it takes another minute to close out my open windows, because each one had an open session and it needs to prompt me on close.
If I open Excel, another minute passes before it loads. If I stupidly decide to change the font for some cells, I expect to see the app lock up for another minute or two. If I want to save the document, I have to go through several layers of menus and garbage just to save to my computer and not the cloud. And then I will still get an alert in Excel that indicates it was unable to save to the cloud and that I need to take some action to resolve the files. The best part is now I can't simply scroll the rows in a spreadsheet in the background while I have another app in the foreground. I have to make another click to bring the spreadsheet into focus to scroll.
If I go into the Azure Portal, I can expect no sane or consistent interface. Components seem to be cargo-cult designed without regard to how they are used. I was looking at some app service logs today, and I'm stuck using their ugly UI, which truncates/hides important data, like the actual log message. The UI is so antithetical to what a logging user interface should look like (think Splunk or Sumo), you'd be better off just downloading the CSV file.
The same goes for build logs in Azure DevOps. It doesn't word-wrap, so now you need to scroll all the way across the screen, line-by-line, to figure out the message. If I click on the Raw Log link, I get the logs, but polluted with ANSI escape codes. If I try to copy the text as is, the UI may or may not select all of it.
And this is just the very start for the disaster that is Azure Portal/DevOps.
And Teams, lol.
When you have to deal with all these problems, all day long, it really starts to sink your productivity. MSFT badly needs a leadership shakeup, but I doubt that will happen as long as Azure is profitable. Until then, I expect to see more decline in the quality of their developer and productivity applications.
Presumably, there were a lot of lifers at MSFT during the Windows 7 years, yet I would consider the software being delivered during that period as being pretty good.
The decline happened after they fired off their SDETs and turned their customers into testers. The hyperfocus on cloud didn't help. Neither did the move towards clunky Electron apps, instead of first-class Windows applications.
SSMS is become unbearable to open. It's insane how long it takes. Not even to connect, just to get the dialogue box to select an instance to connect to.
One idea about Excel: Keep it running all the time, with a new, empty doc. Then, you never pay the start-up cost. Laughably, that is my strategy these days. Same for Adobe Acrobat.
Microsoft has always had shit UX. MS-DOS 6.2x was an improvement over DOS 3.3. Windows 3.x wasn't as clean as it could've been. Windows 95 went too far on special effects. Windows NT was the successor to the failure of the OS/2 joint venture. OS/2 was quite an achievement and very stable. NT 3.1-3.5x had teething pains while NT 4 SP6 became solid. Windows 2000 was quite solid and preferable to XP.
Apple has had functionality that is moved and taken away in the quest for "perfect tomorrow someday" UX and functionality (merging macOS and iPadOS).
No Linux/GNU distro, not even Ubuntu, has reached a point to where functionality AND UX were both good enough for average non-engineer grandma.
FreeBSD usability is about in the Linux/GNU camp. There's less to manage under FreeBSD but it's not as ubiquitous.
> No Linux/GNU distro, not even Ubuntu, has reached a point to where functionality AND UX were both good enough for average non-engineer grandma.
Disagree. I have my (extremely) non-technical wife and 8 year old son on Linux laptops and it's totally fine. Is it perfect? No. But it's good enough. I'm the one who updates them, though.
Support is key: I don't support my (extremely non-technical) mom's Windows laptop or her iPhone because there's nothing for me to do. (My mom had a stroke and is mid 70's.) She wouldn't be able to use anything other than Windows because the retraining burden would be too great. I would expect most 8-year-olds to easily move between any OS because they're far more adaptable than ye olde fuddy-duddy adults.
My elderly parents use a Xubuntu laptop just fine. It ended up with Xubuntu because I got pissed off at the Win 11 Home Edition registration bullshit, and all my parents really use is the web browser.
If I did it again, I'd probably use Linux Mint Cinnamon.
IMO, Windows UI/UX has gotten so bad, even Gnome seems like an drastic improvement.
Sounds like a Chromebook or iPad would've been enough.
I did a T480 motherboard integrated i7-8550U -> discrete i7-8650U swap and rebinding of the license with their phone support. I somehow finagled a Windows 11 Enterprise (retail channel) license that Redmond doesn't have a problem with. It's a cheapskate gaming machine for an eGPU RTX3060 Ti. They can try to pry my dual battery T480 from my cold dead hands, but I'll probably reinstall Qubes on it.
< It ended up with Xubuntu because I got pissed off at the Win 11 Home Edition registration bullshit, and all my parents really use is the web browser.
Yes, a few weeks I spent an hour on the phone walking my Dad through registering Windows 11 and registering his copy of Office. It was abso-fucking-lutely terrible. He didn't even want a live account/address, nor one drive. My Dad browses the web and writes short stories.
Despite spending $149 on his Office license, he has warmed to Google docs, especially that he can share the document and I can help him with the layout very easily. I hope he'll just stop using Word altogether.
WRT updates and restarting, we went through quite a bit of effort tracking down exactly the settings needed to disable update checking, because once it checks for updates it seems like there is basically no way you can prevent it from rebooting on it's own at some point in the future. And this is on Windows Server, where you'd think they'd be even more reluctant to just reboot.
I think this is where MS would tell you to upgrade to an "enterprise" version of Windows Server. These generally have more control over things like automatic updates and rebooting.
I'm neither on Windows nor do I currently write C/C++ stuff, but I just saw a video about RemedyBG, a debugger for Windows that seems to be really nice. Maybe that would help you at least in that regard.
There will always be an option to replace morally bankrupt software... It may come late, but it will come. Napster came out of a need to de-fund greed, so did bittorrent and re-engineered/edited software builds, and many other solutions created by the more ethical/enterprising parts of the underworld.
As shareholder-driven greed consumes these large monolithic companies and their overhead eats them out of their massive office campuses, they will eventually realize that product coexistence and maintaining product reliability/performance while putting customers first will be the only way to survive once their guaranteed government buying partnerships face a new audit and dry up fast.
The cycle is terrible to productivity and for our sanity, and ad revenue is not worth all of the stress they put all of us through. There will also be lawsuits and many other issues raised eventually once people realize that getting robbed at an ATM because you had to watch a 5 minute ad, before your withdrawal completes, is a very stupid idea.
I've been doing IT for quite a while now. About 25ish years ago a bloke called Ben Greer sent me some patches to get 802.1Q VLANs running on Linux which sorted a snag for me on a rather large network with no Windows domain and a lot of workgroups. I whipped up a lot of Samba daemons - one per network per workgroup and combined their browse lists. There was something like 100 of them running on an old Compaq cream coloured tower thing.
A few weeks ago I filed a couple of bugs a few days before buggering off on hols. Both were fixed within a couple of days. One was for a pretty esoteric snag on Arch Linux relating to Python, Kerberos and WinRM. Ansible now works again. I can't remember what the other one was but it was to do with Graylog - again fixed with alacrity.
I've done time on the Gentoo and Arch and many other forums and sometimes actually fixed stuff 8) and of course had stuff fixed for me. I have also been called a wanker and may have had some fairly heated discussions - that is the way of the internet. I have also apologised for being a knobend a few times and that gets a good response.
My experience with Windows is not the same as my Linux experience and I started off with Windows 2. OK I actually started off with a ZX80, ZX81, Spectrum and I still have a C-64 (with USB nowadays).
When you search for a Linux snag these days you get an awful lot of blogs that are at least nicely formatted copies of canonical sources. I use uBlacklist to get rid of the worst. Search for a Windows snag and you generally fall into a whole world of hurt. sfc /scannow or reinstall your OS is a common answer. I'll drop here that sfc has never fixed a snag for me (I worry about roughly thousands of Windows boxes) but someone a third my age will still insist it is the dog's nadgers.
Windows isn't malware, it's just lost the plot. There is a lot to love still in there somewhere but the copy n paste icons are an absolute aberration and W11 is just plain odd in places.
You cannot use telemetry to work out what is going on in someone's mind. My wife uses Arch Linux with KDE as a WM and has no idea what that even means. I divorced her from Windows at Win7 and never looked back.
Great comment. Between sfc /scannow and a format you can also try to revert to an earlier snapshot. Don't forget the ever so optimistic Windows Troubleshooter. I think I've seen it fix a network issue once by restarting the interface. :)
I've just finished updating my work Windows boxes to Server 2019. We are a small firm with three DCs, a FS/PS, ePO (Mcafee/Trellix), RDS/RDG. The ePO at least was 2008 R2 originally. I always do inplace upgrades these days.
If you do a search you will probably find that I am apparently so wrong that I need to report into Room 101.
I think that what goes wrong is an inability to look at logs - laziness. Windows has rather a lot of logging. It took Linux years to catch up with journald (Events) plus various random text logs in /var/log. No matter how much logging you have, getting technicians to actually look at the bloody things is a really hard problem. The little darlings think they are omnipotent and can deploy a cure based on what I call wankery. It might work but probably doesn't.
Medics are way ahead of us here - I've never heard the term contra-indication in IT. Perhaps we should learn from experts in dealing with weird symptoms and very, very complicated diagnosis.
> sfc /scannow or reinstall your OS is a common answer
Microsoft outsourced the support to some chinese guys who are the masters of /dev/null'ing any support request they have. It's always to reboot your machine and sfc /scannow, but of course, it never works.
The classiest part of the act is that the same support guys mark their own "answers" with big green check marks to signify their "importance". I think this helps them to meet the corporate metrics. The whole Windows support thing is a total joke since 2008.
It's not just China. I generally find that responses come from the Indian subcontinent.
I could participate myself but MS insist on using users as unpaid "volunteers" to work for them and that's not my bag. If they'd like my experience then they might feel free to pay for it and I am very cheap compared to the usual 'softies.
In some countries, time spent on MS Social is a requirement or sufficient for a job.
Oh lord: the newspeak conflating "malware" of "they don't give me all of the source and all of their tools" with "ransomware encrypted all pictures of deceased grandma". I don't have a horse in this cockfight between religions because there are strengths and suck in each.
Windows annoys me because of prevalent active app stuttering and lag that has never been resolved with more soft-realtime IO and context schedulers to what the user is focused on. Debugging tools have somewhat improved but you have to be a nuclear engineer to locate what is impacting performance.
Ugh.. not this again. It can't be malware if there is no malicious intent behind it. Not greed, not carelessness, not incompetence not apathy but malice, as in MS must desire to cause harm to its users.
Let's tally up every RCE in GNU software and call it a backdoor if we're misrepresenting other's intentions.
Shame on GNU, the quality of the foundation's work speaks for itself without this drivel.
Why would ANYONE use windows. 40 years of bs crap. Why can’t they just keep windows classic with all the legacy code and build a new light system with no legacy? Has ms had any innovation ever!!!
> Microsoft is imposing its surveillance on the game of Minecraft by requiring every player to open an account on Microsoft's network.
It's ok when it was Mojang, but it's "surveillance" when Microsoft does it? I mean - Mojang required a login account for the longest time, also. I considered it a good approach simplifying accounts so people can play together across iOS and PC.
Anyways... onto commenting the comments below;
I see people complaining about Office being slow, and taking very long to load. Office apps are one of the fastest applications on my system, often taking a second to load. Yes, Excel loads in a ~second. It's so fast, that it puts LibreOffice to shame taking at least 30 seconds to load for me. And Excel is one damn polished of software. I wouldn't trade it for Calc, or Apple's Numbers.
VS being a terrible dev environment; yet I don't see anyone suggesting something better. It takes 15 seconds to load cold and is pretty quick to load after. Doing dev in Notepad++ and then building in VS afterwards is just.. what?
> Free software makes it possible for me (my business, etc) to pay someone other than the original developer to make the software do what I want. No proprietary license gets me that.
Why not pay the original developers if they're doing such a good job?
> Nonfree software is controlled by its developers, which puts them in a position of power over the users
Yet most users cannot modify software because they do not have the technical know-how. The developers are still in control, and even if you are an experienced developer yourself - there's a good chance you aren't familiar or comfortable enough with the target language to make such changes yourself.
I do acknowledge that FOSS has an important place in our society and our future. I've been a UNIX/Linux sysadmin for a very long time, yet treating it as something of a religion and taking a "cult-like" approach isn't really the best way to make progress. GNU is being held back by this, because there is tons and tons of non-free software built by sole developers who don't want their work and time stolen from them (which the GPL purports to prevent- but this is not true in practice). Yet by the FSF definitions, that makes sole developers evil- and pushes many of them away from the free software platform.
Re; forced software updates...
Anyways, forced software updates is the results of years and years of people not wanting to update their software through their own obstinance, getting viruses and malware as a result and blaming Microsoft and leaving them on the hook for damages. Yes, you can disable the updates with Windows Server - I have NO IDEA how some of you guys are complaining about that. I'm only a middling Linux admin but even I know how to manage Windows boxes at a basic level.
> A lot of malware comes from piracy, not from vulns.
My sweet summer child. A lot of malware and/or viruses do not come from piracy. They come from people who click random web links and/or e-mail and run software that pops up on there. While piracy obviously exists and the web has made it much easier to do so, the vast majority of malware and viruses come from "stupid" users that don't understand what it is they're doing.
Anyways, that's all the energy I have for today. I'm not under any illusion about Microsoft either but being evangelistic about this matter isn't going to help anyone.
250 comments
[ 3.0 ms ] story [ 246 ms ] threadCalling Microsoft software malware is so over the top that it turns off normal users and is only appreciated by people who are already committed to free software.
Well...
Joke aside, its not like I came up with this definition.
> Malware is software that is installed on a computer without the user's consent and that performs malicious actions.
https://www.mcafee.com/en-us/antivirus/malware.html#:~:text=....
Do you manually inspect every CPU clock cycle spent, or every data connection or packet sent by every piece of SW running on your system, to make sure it was all done with your consent?
Excel, like the entirety of the titles in the Office Suite, was originally developed on the Mac - where it was hardly used since the Mac wasn't targeted to business at the time. Excel was ported to Windows 2.0, but like the Mac, hardly any business was using that. In the 1980's business ran on Lotus 123, not Excel.
All that changed after the release of Windows 3.0, but by that time Microsoft was already "The Dominator" as you put it. Excel 3.0 was released for Windows, Lotus took way too long to come up with a Windows release and Excel, with its GUI, had come to dominate the market. The rest is history, as they say.
Later we found out that while Microsoft was saying one thing in industry press releases and conferences, they were preparing Windows 3.0. I've never trusted that company since, and they've never failed to meet my low expectations of them.
Thing is, I remember a very different Microsoft. They used to be a great company. Then they made a deal with the devil in the 1987 timeframe and they've been evil ever since.
Lotus had its own proprietary menu structure generally invoked with slash. ALL programs did when Lotus was introduced. Its partisans made it hard for Lotus to abandon it, but its failure to shift into the new reality meant the oodles of new people flocking to computing as Windows exploded went to Excel by default, because 1-2-3 was impenetrable.
Word vs. WordPerfect played out in much the same way.
In both cases the MSFT offering was, in many ways, superior, but a key piece of it was the legacy player's refusal to adjust to the common menus Windows users expected.
"208.3. Microsoft's unresponsiveness and its actions to delay the Windows 95 license negotiations with IBM followed soon after IBM announced that it was seeking to acquire Lotus Development Corp., a direct competitor of Microsoft in messaging and office suite software, among other products.
IBM announced on June 5, 1995 a hostile takeover attempt of Lotus. On June 11, IBM and Lotus reached an agreement for IBM to acquire Lotus. IBM completed the acquisition on approximately July 5, 1995. Norris, 6/7/99am, at 35:9 - 36:12. Among other products, Lotus offered Lotus Notes, an e-mail product, and Lotus SmartSuite, an office productivity suite, that competed with Microsoft's software. Norris, 6/7/99am, at 35:21 - 36:5."
"Microsoft repeatedly penalized IBM for competing against Microsoft
Microsoft withheld a Windows 95 license from IBM until 15 minutes before the product's launch because of IBM's preloading of competing products Microsoft conditioned access to critical marketing support, and other terms and conditions for Windows provided to other OEMs, on IBM's not preloading competing products with the PCs it shipped Microsoft sought to condition substantial MDA price reductions on IBM's ceasing to support competing products
Search for "quicktime" for more joy.
That they did it is clear. Whether they did it for simple profit (Office/Works) or to further enhance the moat around Windows (Java, QuickTime). They did it using Windows as both carrot and stick, to gain compliance. Because OEMs had no other choice.
Excel is genuinely excellent. I was as big a MSFT-hater in the 90s as you are likely to be able to find, but credit where credit is due: Excel provides a level of data analysis power to a lay user that is just not available elsewhere.
Sure, tech people (cough HN readers cough) often want to trot out arguments about how "well, that sort of stuff just shouldn't be done in Excel! They should learn to use R or SQL or some other $MoreCorrectTool". That's poppycock. Excel is already on the analyst's desk. It's immensely powerful while ALSO being very accessible. Nothing else comes close.
Excel was just better by miles.
I think building a good spreadsheet program that is both accessible and powerful is very difficult. Excel might be the most non-trivially used "programming language" in the world.
It is undeniable that Excel has provided massive value for end users and been massively empowering for them.
Whether or not Microsoft software is malware is a different question that if Microsoft is a good or bad company.
Excel rode on the popularity of VisiCalc and Lotus 123 for years.
https://thehistoryofcomputing.net/visicalc-excel-and-the-ris...
“They released an early spreadsheet tool called Multiple [sic] in 1982. But Lotus 1-2-3 was the first killer application for the PC.”
- https://www.gnu.org/philosophy/selling.html
- https://www.gnu.org/philosophy/selling-exceptions.html
This is moronic pseudo-religious doublespeak. They're saying it's acceptable to charge a fee for the one part of software creation that's effectively cost-free.
Even most of the vendors of free software today (e.g. Confluent) are using some kind of SaaS or open core model.
Unlike physical objects, where the sale accounts for the cost of production of material etc, what value does someone trying to sell a GPLv3 software add? I argue that the added value is zero, and hence, the price must be zero as well.
At the time, I thought it seemed a little over-the-top.
Yet since then, I have personally lost control of many of the computing devices I use. The phone was the biggest one, and I can't (don't?) avoid phones altogether like RMS. His words were actually accurate.
Now, I do think that some people try to get a message across, and are not heard, so they use stronger words until they get a response. Those messages do become over-the-top.
But I have come to realize that RMS himself carefully chooses his words.
No, you haven't. Or rather, you didn't lose control because of any software license. We lost control because of cryptography and enforced signature verification. Just because software is licensed as open source doesn't mean it allows for third-party firmware modification.
Maybe it's because I came up through the RE scene, but the false association between open source and ownership or security has always stood out to me. Many of the modding and hacking communities that orbit proprietary software are/were more vibrant than anything that exists in the totally open source space.
I'm not sure what you mean by this. It's the law that's a barrier, not cryptography. If you break the DRM on a game you can still face legal consequences, even if the company that licenses it doesn't support it anymore.
Exactly, you just named a subset of all the things you could do to a machine, partitioned by copyright protection. I'd agree with your general point in the context of power users, but it's worth noting that the average Joe will try to faithfully follow company policy as well as the law.
To run afoul of the DMCA a modder needs to (1) strip a binary of its copyright protection (2) distribute the unpacked executable or the unpacker tool (3) likely facilitate mass third-party piracy (more damages = more lawsuit potential). It’s principally warez and hacking groups that meet these criteria, and they probably ought to be sued.
I’ll add a point in your favor and say that for non-games DRM is often a big problem when reversing or patching code.
Yes, the term for this is "tivoization." It is a large part of why the GPLv3 exists.
> Just because software is licensed as open source doesn't mean it allows for third-party firmware modification.
Yep, "open source" and many copyleft licenses are insufficient for this job.
> “Installation Information” for a User Product means any methods, procedures, authorization keys, or other information required to install and execute modified versions of a covered work in that User Product from a modified version of its Corresponding Source. The information must suffice to ensure that the continued functioning of the modified object code is in no case prevented or interfered with solely because modification has been made.
> If you convey an object code work under this section in, or with, or specifically for use in, a User Product, and the conveying occurs as part of a transaction in which the right of possession and use of the User Product is transferred to the recipient in perpetuity or for a fixed term (regardless of how the transaction is characterized), the Corresponding Source conveyed under this section must be accompanied by the Installation Information. But this requirement does not apply if neither you nor any third party retains the ability to install modified object code on the User Product (for example, the work has been installed in ROM).
(https://www.gnu.org/licenses/gpl-faq.html#GiveUpKeys clarifies that this applies to cryptographic keys.)
lol. Look, in practical terms - as well as by definition - proprietary software is owned by someone else.
It makes them feel righteous.
but they live in their own world
For example, (referencing the "Jails" section here), the author says, "Jails are systems that impose censorship on application programs." Not sure if I'm interpreting this next quote from the GNU.org site (https://www.gnu.org/distros/free-distros.html) correctly, but it seems like this is exactly their definition of a "Free GNU/Linux distribution" is meant to do: "They will reject nonfree applications, nonfree programming platforms, nonfree drivers, nonfree firmware "blobs," nonfree games, and any other nonfree software, as well as nonfree manuals or documentation."
"They will reject" seems an awful like "won't run on my distribution", and not because it's not compatible because of runtime or kernel differences, but because of its 'being nonfree'.
From Merriam & Webster:
BIGOT: a person who is obstinately or intolerantly devoted to his or her own opinions and prejudices.
I love free software. I don't have to pay for it, I can share, customize, and use it as I want without the author of the software ever needing to know or care, because he did it out of either love for software or because he had a need to fulfill and was nice enough to release it to the world.
I also use paid software and software containing "nonfree" software blobs. I have an Android because it works for me. I use Spotify on all my devices, and pay for Premium, because I find it convenient, the same reason I use GNOME and VS Code and drive a used Chevy.
I value my time and quality of life too much to be extreme "beCaUSE iT wASN't FreEEE".
Just my 2 cents. I'll have that with a side of Kale please.
Cinnamon DE is great.
i3wm or ratpoison tho frfr
plasma DE is really nice though; and I use that on my artix install. Love it. It's like windows but it doesn't suck
Plasma understands that; as well as PLENTY OF OTHER DEs but your casual linux observer just thinks "wow gnome looks ugly" and it does, it always has, i still have absolutely no idea why anyone would roll a distro out with that... mess.
[0]: https://survey.stackoverflow.co/2022/#section-most-popular-t...
most game developers are on windows
whole tech sectors have been captured successful by Microsoft in the past and most developers there have to use Windows weather they like it or not, some of them are _really_ big (but also not super vocal or visible on GitHub as they tend to be quite proprietary in all of their actions).
etc. etc.
Idk. how realistic 75% is (e.g. what it counts as developer who/how people where asked etc.) but it's probably a high rate anyway.
EDIT: Also also probably using one doesn't exclude the other and doesn't necessary exclude them running it in a VM ;=)
* Word
* Excel
* Honestly, Teams
* Exchange / Outlook, for group scheduling
MS software I use because I have the most experience with it, and am thus more productive than I'd be with something else:
* SQL Server
I mean Microsoft Word, the word processing tool that runs as a native app on one's computer. The degree to which the online or Teams versions do not play nice with the "real" one is a HUGE own-goal by MSFT and reason enough to avoid them if you can. If you're going to have an online tool called Word, it damn well needs to work EXACTLY LIKE THE REAL VERSION.
Now where it all goes pear-shaped, is not the technical side of it all, it's the UX and marketing teams that got involved. One big example is the "save" dialog that's been butchered beyond describing (even in the standalone desktop version of Word) in order to promote OneDrive and other microsoft offerings.
Downloading a lot of files off Teams usually fails and gives a load of html files.
The ‘integration’ nag for one drive which can’t be dismissed or enabled but seems to work.
It’s an absolute mess and I’m very surprised to encounter someone who finds it works well.
Are you sure?
We are a ~15 person company and there isn't a (work) day where someone doesn't have problems with teams. I can't say for sure because I hadn't used it a lot but jitsi seemed way better and google while having far worse UX was more reliable. Like screen sharing stopping working (without any error or anything just stand bild) during a meeting is pretty common, or abstruse things like one time the chat was just missing until I restarted it or another time there was a partition of rooms until everyone left the meeting and then reentered it again.
Similar problems with outlook are also not that uncommon (but more rare), like e.g. calendars not syncing, the only things really speaking for it (compared to some other good team calendar applications) IMHO is it's integration with teams... which is worth something, I mean it's pretty convenient.
* Calling each other;
* Ad-hoc meetings internally, because our normal meeting platform has become cumbersome (GoTo);
* IMs
I see it'll do other stuff, but we're not really using those aspects. And I should also be super clear that I'm not saying Teams is actually GOOD. I have just found it to be better than any of the alternatives I've seen or used.
That's also very much how I feel about Outlook/Exchange. I mean, look, I'm a Mac guy. I use the Mac mail client almost but not quite exclusively; I also have an RD session up to a Windows VM all day every day for other reasons, and run Outlook there. The gap that drives me to Outlook/Exchange is scheduling. It just works, and works in a way that nothing else I've seen approaches -- even when messaging people outside our org.
I wish Apple calendaring was this good. I really do. But it's not.
No it doesn't. You are free to install whatever software you want. None of these systems will prevent a user from using non-free software. They just won't encourage it.
I find GNU's position too dogmatic, but it's not as bad as you pretend.
Do you value the time and quality of life of the billions who can't drive a Chevy, nor will ever have a device that can run an Electron app or $10/month for Spotify?
It's not all free as in beer you know, free software is meant to serve a free society by allowing everyone to participate in the social construction of technology, not provide consumer luxuries. I think the term you're looking for is "open source"
> proportion of women represented in the computer science field peaked around the mid-1980s and has declined ever since
I thought ‘surely not’. I was wrong.
https://www.scientificamerican.com/article/there-are-too-few...
Steve Jobs was in a way quite typical of the sort of people who showed up in the IT field. He would have a tough time getting hired today.
I suggest staring deeply into why culture and natural language work like they do (or used to do, seems only stubborn 'artistic' types are working with culture like that nowadays)
That's certainly one common recent use of that term. Although more often than not it's just "I don't like them but don't really have argument that passes first smelling therefore they are bigots"
It's not; first (as referred in article) is preventing user from using the machine as they see fit.
Second one is not accepting software into distribution that tries to take user's freedom to modify it; they don't stop you installing non-free software in free distribution, they are just not shipping the non-free software. It's a stance on not promoting (by including and making installing trivial) the non-free software but you are still free to install on your own.
Going by tired
> "They will reject" seems an awful like "won't run on my distribution", and not because it's not compatible because of runtime or kernel differences, but because of its 'being nonfree'.
Nope, it explicitly means "will not distribute, with their distribution, for free". Software actively stopping you from running other software would actually be nonfree, as the whole deal with OSS is "do with sofware what you want" and GNU deal being "and pass same priviledge forward"
It doesn't sound like "won't run on my distribution" to me. It seems that you intentionally took this line out of context just for the argument (or to defend MS).
The full paragraph:
> These distros are ready-to-use full systems whose developers have made a commitment to follow the Free System Distribution Guidelines. This means they will include, and propose, exclusively free software. They will reject nonfree applications, nonfree programming platforms, nonfree drivers, nonfree firmware “blobs,” nonfree games, and any other nonfree software, as well as nonfree manuals or documentation.
Next paragraph:
> If one of these distros ever does include or propose anything nonfree, that must have happened by mistake...
"This means they will include...". Not "This means they will run...".
"include or propose...". Not "execute or install...".
It clearly means that they don't include nonfree stuff with their systems. Not they don't allow nonfree stuff to run.
As an obvious example, I honestly can't imagine the Linux kernel having had the impact that it has had if all of the myriad corporations building on it weren't forced to contribute their work back to it under the same conditions.
> The paradox of tolerance states that if a society is tolerant without limit, its ability to be tolerant is eventually seized or destroyed by the intolerant.
On the side, I've even learned to like kale on pizza. The Kale-Zer Soze at Central Pizza in Seattle is outstanding.
Arch is a swing and a miss
> Arch has no policy against distributing nonfree software through their normal channels, and nonfree blobs are shipped with their kernel, Linux.
Debian is a no go
> Debian also maintains a repository of nonfree software. According to the project, this software is “not part of the Debian system,” but the repository is hosted on many of the project's main servers, and people can readily find these nonfree packages by browsing Debian's online package database and its wiki.
Sorry Gentoo
> Gentoo includes installation recipes for a number of nonfree programs in its primary package system.
BSD
> FreeBSD, NetBSD, and OpenBSD all include instructions for obtaining nonfree programs in their ports system. In addition, their kernels include nonfree firmware blobs.
Nonfree drivers are not OK.
Nonfree BIOS/EFI programs are not OK.
Nonfree kernel blobs are not OK.
We may use them out of necessity and pragmatism (some nonfree software is better than 100% nonfree software! an Arch user still has more intact freedoms than a Windows user), but we should still reject them ideologically and remember that they are tolerated but not accepted.
The same also applies to copilot. I've trained myself to give it prompts to generate code that matches what I would've generated by hand years ago. But without copilot, I wouldn't program. Simply not practical.
Why is it unethical?
Would it be unethical to be required to use an entirely mechanical machine that was still covered by patents or trade secrets?
The idea that non-free software is unethical and should be abolished is the bedrock of all of the FSF’s philosophy, and is held almost as a religious view for RMS and his adherents.
All these questions are answered in book "Free Software, Free Society" (with better explanations than I can give).
> mechanical machine that was still covered by patents or trade secrets
I would say, yes, it would be unethical, too.
> people can readily find these nonfree packages by browsing Debian's online package database and its wiki.
Today, I am using Debian myself and avoid nonfree packages knowingly. However I am glad that FSF keeps reminding users about it. Note that they do not prohibit the usage of Debian. They just don't endorse it.
Why? If a vendor sells a fingerprint scanner or something why do they need the driver to be closed source? Why does somebody tweaking the source bother them?
I've never really understood this. Sure, Microsoft word needs to be closed source but why a driver?
Plus there's just a lot of belief in the nonfree circles that the software freedom movement is just religious zealot hippies who think software should never be paid for and that microsoft programmers should work for free and starve, and that anyone who does develop novel software should have it immediately ripped off and copied by competitors and make no money because capitalism is bad.
(I'm being a bit hyperbolic but not much.)
If I fix a bug or desire some new functionality and implement it for myself, am I free if I cannot legally share it with my neighbors? Certainly not.
Also depending on exactly how they release the source, you may still not know that's what the blob is actually doing. I suspect many vendors, even if forced somehow to provide source, would try to not allow (or make very annoying) building it or signing it. And if you can't build it yourself, or if you can but can't run it and the build isn't bit-perfect reproducible, then you can't know the blob is actually the output of building the source.
You could decompile but let's be real, if security researchers looked at Heartbleed not once but twice, specifically in the context of it being a bug, and both times asking if it's a security vulnerability, and said no - then I am not going to believe that a human inspecting decompiled machine code is going to be very good at finding well-hidden malware in blobs.
In the ports tree, there was a makefile. http://cvsweb.openbsd.org/ports/devel/jdk/1.3-linux/Attic/Ma...
If you open that makefile, there is a url. http://java.sun.com/j2se/1.3/download.html
If you open that url in a browser, you will be presented with the opportunity to download nonfree software.
I no longer need to worry about Windows, but now wondering what to do about UEFI. I've heard about things like coreboot, but doesn't seem to be offered by mainstream manufacturers. Not sure I wanna risk bricking my machine. Has there been any recent progress here?
If you want to be fully free, you may be interested in the Raptor Talos II, which is based on OpenPOWER and has free firmware. It is pretty pricey however.
It's shocking to spend time with Federal Workers and see how much of their days are wasted waiting for anti-virus and other bad closed source software, and then to realize we all are paying taxes to support this! Same goes for Oracle. BG and LarryE got the country to fund their yachts and stuck us with the bill.
I'm gonna run for Congress under a platform of banning any Federal money from being spent on closed source, non public domain software.
The ideal user to Microsoft appears to be an uninformed one, not interested in control and apprently willing to reliquinsh it. Not all Microsoft users are uninformed or disinterested in control. There are profound exceptions. Some of those exceptions will read this comment and respond accordingly. But hundreds of millions of non-exceptions will remain uninformed and disinterested, to the benefit and delight of Redmond.
Obviously, malware can thrive under the auspices of uninformed, oblivious users, Microsoft's ideal customers. Microsoft will purportedly protect them, when it wants to and assuming it can; of course, it won't protect them from Microsoft itself. For those Microsoft employees and users not reading, commenting and/or voting on HN, enjoy the ads injected into Windows. Microsoft is happy with the result.
1. Worth noting perhaps, Microsoft was making an estimated $2B annually from patent royalties on Android. https://finance.yahoo.com/news/microsoft-may-relinquishing-b...
(The point that commenters/voters appear to be missing is that it is the nature of the customer/user that creates the attractant for malware as much if not more than the software itself or the size of the user base. GNU regardless of what one thinks of them is AFAICT advocating for user control in this communication. That implies users who care about control, not the average Microsoft user. Users who are paying attention and who do not blindly trust companies like Microsoft. One can argue it is the "insecurity" of the software or the number of users that creates the attractant, and there is certainly some truth there, but it makes no sense if, e.g., the user base is a large number of paranoid users who read HN. They are going to avoid/prevent/detect malware better than a random sample of Microsoft users. User behaviour is relevant. If 9 out of 10 users who run program X on operating system Y are careless, then program X and/or operating system Y is arguably an excellent target for malware.)
Were that king Linux or Mac or BSD or god damn Haiku, they would be seeing the brunt of malware attacks instead.
If MS dealt with all their technical debt and built a brand new OS following proper disclosure as well as OSing most of their kernel, you'd see that it would be much harder to exploit.
Think of linux that runs over 90% of all servers in the world. How many ransomware attacks have they been hit with?
Calling software malicous because it prioritized security over freedom just reads to me as being ignorant.
> Pick your poison
Alternatively, you could use a GNU/Linux on PC and phone.
- Google: https://www.gnu.org/proprietary/malware-google.html
- Apple: https://www.gnu.org/proprietary/malware-apple.html
I can figure out iPhony is an iPhone, but I'm not certain what iBad and iMonster are supposed to reference. I could dig into the source references and try to figure out WTF they're trying to convey, but that shouldn't be necessary.
Things like this hinder their credibility.
Regardless of context, it harms one's argument (and is blatantly disrespectful) to refer to people or groups or items with labels other than their actual names. That kind of foolishness has the same level of rhetorical value as MAGA types calling the opposition party the "Dummycrats" or whatever.
The implication is that FOSS developers have no control over their users. Sure, you COULD fork your favorite tool's repository and spend many hours trying to figure out how to make it do 'what you want instead of what the developers wanted'; but very few people are really willing and/or able to do that.
Nearly all software (open source or proprietary) except trivial programs, are controlled by the developers who build it and maintain it. They are given that power by the mere fact that most programs are sufficiently complex that few others can and will assume control instead. If someone does (and that someone isn't you) then the new boss is same as the old boss (to quote a line from a song by The Who).
Free software makes it possible for me (my business, etc) to pay someone other than the original developer to make the software do what I want. No proprietary license gets me that.
In the context of personal use software by an individual it may not make sense to pay, but in the context of groups of individuals or a business it becomes viable.
Would also be appreciated to see an example.
And this does happen! The response to Oracle's anti-F/OSS, user-hostile moves when they bought Sun is a powerful example.
Free software also means that you can pay developers to take the software in a different direction if you can't do so yourself.
There are lots of user-serving forks in the world. Cinnamon and MATE really did serve Gnome 2 users who were pissed off and disappointed about what they saw in GNOME 3. LEDE really did resurrect OpenWrt by giving people a way to push the project forward when development had stagnated. yt-dlp saved a bunch of youtube-dl users who would have been left in the lurch.
Of course creating and maintaining a fork requires non-trivial resources. But forks do happen and their possibility absolutely makes a difference to end users, lots of the time.
Few people are willing to do it because the kinds of restrictions/annoyances imposed by free and open source software are much less severe than that of nonfree software. Sure, there might be the odd feature or two missing in a FOSS project that you would like but that the author blocks from being added to the project for XYZ reasons. I'll concede that those kind of decisions give the maintainer of a FOSS project some semblence of "power". But it's not like proprietary software, where trackers, atrocious adverts and news links (looking at you, Microsoft start menu) and other arbitrary restrictions are a continuous reality. If those things were in a FOSS project, you can bet that _someone_ would make a fork with that stuff removed.
I’ve also contributed patches upstream to open source software plenty of times and the majority of time the maintainers have been willing to accept it.
Tons of Linux distros maintain their own set of patches on top of loads of software to make it work better with the rest of their system.
Sure, there are examples where it’s infeasible to change things but in practice it’s a lot more likely than with proprietary software.
I was right.
And yet I hate what they're doing now. I turned off automatic updates, or so I thought, and returned to my machine one day last week to find it had restarted anyway. I'd spent a fair amount of preparing a commit and most of my work had vanished. Another time I returned to an unwelcome restart and I had actually lost a fair amount of work. Sure, I retyped it, but really?
On the same machine, I have another account on the same machine that I use for testing. Every single time I sign in to that account, I have to start Task Manager as an administrator and then kill Windows Compatibility Telemetry, even though I have allegedly(?) disabled it several times. Otherwise it consumes so much CPU that I can't get my work done.
Also, Visual Studio is a hot mess. Why do I have to wait when I drag and drop an int into the watch window? Why does my machine slow to a crawl when I'm running debug builds? Why are there so many bugs with breakpoints? A lot of the time, I have to use page up/page down to see the breakpoint appear even though I hit F9. Why is it agonizingly slow to type things when I'm debugging? Why do I fantasize about finding a new job that doesn't require me to use Visual Studio? Visual Studio has become my least favorite piece of software. There are days when I just write code in Notepad++ and only use Visual Studio to build.
It's frustrating. The seeming inability to exert developer-level control over Windows makes me feel like my devices aren't mine. For fuck's sake, I had to tell Windows that my internet connection was metered in order to get them to stop with the updates.
Yes, the article might be hyperbolic, but the whole jail argument is starting to line up pretty well with reality. Dear Microsoft, please stop fucking with everything.
Like AI, until someone shows me something that can come up with novel solutions to deeply technical problems, simple tools and intelligence on the part of the programmer is what works.
Unfortunately I don't have any choice in the matter of compiler or platforms. I do some work work on Linux and I use GCC there. It feels much more sane.
Visual Studio makes me want to take more blood pressure medication.
Until then, enjoy your MSFT environment that they've created for the lowest common denominator. You try the code-gen / no-code tools yet? Maybe they'll make you happier. Developers, developers, developers!!
Any power user would be well served to reuse one of their old 7/8 Pro licenses or buy a 10/11 Pro license instead of cheaping out on a Home license.
Also, the user should be able to assume that there aren't multiple switches for the same action. As a general rule - I know this is Windows.
So yes, you can rely on Windows and Microsoft to respect Group Policy. The catch is you need to pay for enterprise privileges if you're not.
I have to imagine there a bunch of software engineers at Microsoft who have worked their asses off and are proud of what the core of Windows has become but are also super frustrated at all of the BS that is loaded on top in 99% of installs.....
Judging from what I've seen from Windows game programming streamers like Casey Muratori and Jonathan Blow, it's not an uncommon practice to live in your preferred editor then flip to VS for build/debug.
It all seems very strange and ridiculous to me coming from a gnu/linux/vim/screen/make/gdb background. Obviously not so much having separate tools, but more the despite having a massive beast of an IDE purportedly providing All The Things, you're still context switching to/from your usable editor.
Those streamers also often resort to using RemedyBG for debugging instead of using VS's integrated debugging. They're basically living without an IDE, while paying all the overhead for an IDE, because the IDE is such trash.
Microsoft helped lift me my family out of poverty. I grew up in Seattle, in the shadow of Microsoft and their Redmond campus. My father was a laborer who helped built Bill Gate's mansion, and who helped built parts of their Redmond campus.
He was incredibly, incredibly, proud, that I drove to work everyday down the streets he had helped plant trees on. He told almost everyone he met that his son worked at Microsoft.
Microsoft did a lot of good in the 90s and early 2000s[0], they created a standard platform which the world built upon. They created a stable foundation for thousands upon thousands of developers to build their livelihood. They paid for generations of children to learn programming in free or next to free programs across the country (and across the world). Since their missteps and the rise of mobile, the industry hasn't had a stable development platform to build upon, and we've collectively wasted untold billions of dollars in productivity on crappy cross platform solutions.
When I was at Microsoft, the customer was king. I put a stop to plenty of bad behavior by saying "is this what is right for the customer?" and if the answer was no, code didn't ship.
But Microsoft is a fear driven company, they were afraid of the rise of the tablet so they made Windows 8, and then they became afraid of Google so they constantly attempt to go for ads. (First they went for search, but then eventually realized Google isn't a search company...) Heck the original reason Explorer was supposed to get tabs was so ads could be shown in new explorer tabs[1].
The other problem is, Microsoft got rid of Software Development Engineers in Test, and they replaced them with devs doing test. But at Microsoft, SDETs served as the last line of defense for the users, as an SDET it was made clear to me from day 1 of my employment that I had the power to stop shipment of software if I felt the code wasn't up to snuff. As a junior SDET I only ever used that power once, I was damn well scared, and when I spoke up and said we had open bugs that I personally considered ship blockers, the lead dev said no problem, the release would be delayed until his team fixed the bugs.
No one has that role anymore. Before I left MS around 2016 or so, I saw a fellow developer instructed by the design team to make a UI component behave as if it followed the MacOS design guidelines. The (junior) dev didn't realize that was what he had been told to do, and it obviously took a long time for him to beat the Windows UI controls into shape until they worked like the MacOS UI controls, and when I asked him why he was doing this, he said he was too scared to push back.
That is what SDETs were for. To hold the bloody line. Facebook's "move fast an break things" motto gave the industry an excuse to gut dedicated test teams, and now all software everywhere sucks, but no one under 30 realizes anything is wrong because from their perspective things have always been shitty and broken.
[0] Yes yes and Linux bad lawsuits etc etc, separate discussion. [1] The ads were to be for Microsoft Store apps IIRC.
I worked for almost 27 years, for a Japanese optical corporation that is synonymous with "Quality."
Their QC department was the most powerful one in the corporation, and all the devs lived in fear of failing QCT, as that would have C-Suite-level notice.
In the US, there used to be a joke, saying "If you have 'Quality' in your job title, your career is over." At this company, it meant you were just short of a God.
That said, it tended to make life stressful for us. As an American development team, we were often treated as "Bad Software Cowboys." Most folks here, would consider us to be completely uptight, anal-retentive, risk-averse, babies, but they thought of us as reckless.
I would welcome this as a cultural change in the US. It just sucks living downstream of people dumping tons of cowboy code into the river.
If I open SSMS, it also takes upwards of a minute to load. Some things I used to take for granted, like being able to see the cursor when scrolling with the arrow keys, instead makes the cursor invisible, so now I have no idea where my cursor lands when I release the key. When a connection to a DB bonks out (which happens at least once a day), it takes another minute to close out my open windows, because each one had an open session and it needs to prompt me on close.
If I open Excel, another minute passes before it loads. If I stupidly decide to change the font for some cells, I expect to see the app lock up for another minute or two. If I want to save the document, I have to go through several layers of menus and garbage just to save to my computer and not the cloud. And then I will still get an alert in Excel that indicates it was unable to save to the cloud and that I need to take some action to resolve the files. The best part is now I can't simply scroll the rows in a spreadsheet in the background while I have another app in the foreground. I have to make another click to bring the spreadsheet into focus to scroll.
If I go into the Azure Portal, I can expect no sane or consistent interface. Components seem to be cargo-cult designed without regard to how they are used. I was looking at some app service logs today, and I'm stuck using their ugly UI, which truncates/hides important data, like the actual log message. The UI is so antithetical to what a logging user interface should look like (think Splunk or Sumo), you'd be better off just downloading the CSV file.
The same goes for build logs in Azure DevOps. It doesn't word-wrap, so now you need to scroll all the way across the screen, line-by-line, to figure out the message. If I click on the Raw Log link, I get the logs, but polluted with ANSI escape codes. If I try to copy the text as is, the UI may or may not select all of it.
And this is just the very start for the disaster that is Azure Portal/DevOps.
And Teams, lol.
When you have to deal with all these problems, all day long, it really starts to sink your productivity. MSFT badly needs a leadership shakeup, but I doubt that will happen as long as Azure is profitable. Until then, I expect to see more decline in the quality of their developer and productivity applications.
The decline happened after they fired off their SDETs and turned their customers into testers. The hyperfocus on cloud didn't help. Neither did the move towards clunky Electron apps, instead of first-class Windows applications.
Lifers may be there still, but they have to march to the beat of the stock market.
Apple has had functionality that is moved and taken away in the quest for "perfect tomorrow someday" UX and functionality (merging macOS and iPadOS).
No Linux/GNU distro, not even Ubuntu, has reached a point to where functionality AND UX were both good enough for average non-engineer grandma.
FreeBSD usability is about in the Linux/GNU camp. There's less to manage under FreeBSD but it's not as ubiquitous.
Disagree. I have my (extremely) non-technical wife and 8 year old son on Linux laptops and it's totally fine. Is it perfect? No. But it's good enough. I'm the one who updates them, though.
PS: Which Linux distro?
If I did it again, I'd probably use Linux Mint Cinnamon.
IMO, Windows UI/UX has gotten so bad, even Gnome seems like an drastic improvement.
I did a T480 motherboard integrated i7-8550U -> discrete i7-8650U swap and rebinding of the license with their phone support. I somehow finagled a Windows 11 Enterprise (retail channel) license that Redmond doesn't have a problem with. It's a cheapskate gaming machine for an eGPU RTX3060 Ti. They can try to pry my dual battery T480 from my cold dead hands, but I'll probably reinstall Qubes on it.
Yes, a few weeks I spent an hour on the phone walking my Dad through registering Windows 11 and registering his copy of Office. It was abso-fucking-lutely terrible. He didn't even want a live account/address, nor one drive. My Dad browses the web and writes short stories.
Despite spending $149 on his Office license, he has warmed to Google docs, especially that he can share the document and I can help him with the layout very easily. I hope he'll just stop using Word altogether.
We ended up with this, BTW:
That is absolutely insane having Windows Server auto update and reboot.
What if it was something critical, like a production line or a server supporting medical devices?
The video that I saw: https://www.youtube.com/watch?v=r9eQth4Q5jg
As shareholder-driven greed consumes these large monolithic companies and their overhead eats them out of their massive office campuses, they will eventually realize that product coexistence and maintaining product reliability/performance while putting customers first will be the only way to survive once their guaranteed government buying partnerships face a new audit and dry up fast.
The cycle is terrible to productivity and for our sanity, and ad revenue is not worth all of the stress they put all of us through. There will also be lawsuits and many other issues raised eventually once people realize that getting robbed at an ATM because you had to watch a 5 minute ad, before your withdrawal completes, is a very stupid idea.
Nice tip. Any downsides to this?
And it's good enough for many purposes.
A few weeks ago I filed a couple of bugs a few days before buggering off on hols. Both were fixed within a couple of days. One was for a pretty esoteric snag on Arch Linux relating to Python, Kerberos and WinRM. Ansible now works again. I can't remember what the other one was but it was to do with Graylog - again fixed with alacrity.
I've done time on the Gentoo and Arch and many other forums and sometimes actually fixed stuff 8) and of course had stuff fixed for me. I have also been called a wanker and may have had some fairly heated discussions - that is the way of the internet. I have also apologised for being a knobend a few times and that gets a good response.
My experience with Windows is not the same as my Linux experience and I started off with Windows 2. OK I actually started off with a ZX80, ZX81, Spectrum and I still have a C-64 (with USB nowadays).
When you search for a Linux snag these days you get an awful lot of blogs that are at least nicely formatted copies of canonical sources. I use uBlacklist to get rid of the worst. Search for a Windows snag and you generally fall into a whole world of hurt. sfc /scannow or reinstall your OS is a common answer. I'll drop here that sfc has never fixed a snag for me (I worry about roughly thousands of Windows boxes) but someone a third my age will still insist it is the dog's nadgers.
Windows isn't malware, it's just lost the plot. There is a lot to love still in there somewhere but the copy n paste icons are an absolute aberration and W11 is just plain odd in places.
You cannot use telemetry to work out what is going on in someone's mind. My wife uses Arch Linux with KDE as a WM and has no idea what that even means. I divorced her from Windows at Win7 and never looked back.
If you do a search you will probably find that I am apparently so wrong that I need to report into Room 101.
I think that what goes wrong is an inability to look at logs - laziness. Windows has rather a lot of logging. It took Linux years to catch up with journald (Events) plus various random text logs in /var/log. No matter how much logging you have, getting technicians to actually look at the bloody things is a really hard problem. The little darlings think they are omnipotent and can deploy a cure based on what I call wankery. It might work but probably doesn't.
Medics are way ahead of us here - I've never heard the term contra-indication in IT. Perhaps we should learn from experts in dealing with weird symptoms and very, very complicated diagnosis.
Microsoft outsourced the support to some chinese guys who are the masters of /dev/null'ing any support request they have. It's always to reboot your machine and sfc /scannow, but of course, it never works.
The classiest part of the act is that the same support guys mark their own "answers" with big green check marks to signify their "importance". I think this helps them to meet the corporate metrics. The whole Windows support thing is a total joke since 2008.
I could participate myself but MS insist on using users as unpaid "volunteers" to work for them and that's not my bag. If they'd like my experience then they might feel free to pay for it and I am very cheap compared to the usual 'softies.
In some countries, time spent on MS Social is a requirement or sufficient for a job.
Windows annoys me because of prevalent active app stuttering and lag that has never been resolved with more soft-realtime IO and context schedulers to what the user is focused on. Debugging tools have somewhat improved but you have to be a nuclear engineer to locate what is impacting performance.
Let's tally up every RCE in GNU software and call it a backdoor if we're misrepresenting other's intentions.
Shame on GNU, the quality of the foundation's work speaks for itself without this drivel.
Anything can be "malware" if you get to make up your own definition for the word.
I've checked the normal places for definitions of malware, and none of them are as broad and simplistic as this.
One of the more interesting additions to the article linked in the OP refers to this site: https://www.techspot.com/news/97535-windows-11-spyware-machi... - I'm not sure if I would consider techspot is good journaling. That's all I'll say about that.
> Microsoft is imposing its surveillance on the game of Minecraft by requiring every player to open an account on Microsoft's network.
It's ok when it was Mojang, but it's "surveillance" when Microsoft does it? I mean - Mojang required a login account for the longest time, also. I considered it a good approach simplifying accounts so people can play together across iOS and PC.
Anyways... onto commenting the comments below;
I see people complaining about Office being slow, and taking very long to load. Office apps are one of the fastest applications on my system, often taking a second to load. Yes, Excel loads in a ~second. It's so fast, that it puts LibreOffice to shame taking at least 30 seconds to load for me. And Excel is one damn polished of software. I wouldn't trade it for Calc, or Apple's Numbers.
VS being a terrible dev environment; yet I don't see anyone suggesting something better. It takes 15 seconds to load cold and is pretty quick to load after. Doing dev in Notepad++ and then building in VS afterwards is just.. what?
> Free software makes it possible for me (my business, etc) to pay someone other than the original developer to make the software do what I want. No proprietary license gets me that.
Why not pay the original developers if they're doing such a good job?
> Nonfree software is controlled by its developers, which puts them in a position of power over the users
Yet most users cannot modify software because they do not have the technical know-how. The developers are still in control, and even if you are an experienced developer yourself - there's a good chance you aren't familiar or comfortable enough with the target language to make such changes yourself.
I do acknowledge that FOSS has an important place in our society and our future. I've been a UNIX/Linux sysadmin for a very long time, yet treating it as something of a religion and taking a "cult-like" approach isn't really the best way to make progress. GNU is being held back by this, because there is tons and tons of non-free software built by sole developers who don't want their work and time stolen from them (which the GPL purports to prevent- but this is not true in practice). Yet by the FSF definitions, that makes sole developers evil- and pushes many of them away from the free software platform.
Re; forced software updates...
Anyways, forced software updates is the results of years and years of people not wanting to update their software through their own obstinance, getting viruses and malware as a result and blaming Microsoft and leaving them on the hook for damages. Yes, you can disable the updates with Windows Server - I have NO IDEA how some of you guys are complaining about that. I'm only a middling Linux admin but even I know how to manage Windows boxes at a basic level.
> A lot of malware comes from piracy, not from vulns.
My sweet summer child. A lot of malware and/or viruses do not come from piracy. They come from people who click random web links and/or e-mail and run software that pops up on there. While piracy obviously exists and the web has made it much easier to do so, the vast majority of malware and viruses come from "stupid" users that don't understand what it is they're doing.
Anyways, that's all the energy I have for today. I'm not under any illusion about Microsoft either but being evangelistic about this matter isn't going to help anyone.