Windows 11 is dying....

[Cyclotron]

Corporate IT wants ease of management and a fleet as secure as possible.

I don't care about what corporate IT wants and so far I have never been fired. Probably I'm lucky...
But I agree that many people are using windows because they have to.

If you're able to operate in your corporate world without being required to comply, you're lucky.  It seems everyone I work with has some compliance requirements, and if they are audited, they have real problems if there's non-compliance.

I don't think people are using Windows because they "have to".  Corporate environments often offer macOS or Windows, but Windows remains the number one choice. In my past two companies, that was the case. People can easily buy Windows, macOS, or ChromeOS, and Windows is the choice of the majority.

It doesn't make sense to me to frame this as if Windows is forced onto people. Why people choose Windows may be myriad, but they choose it over macOS and ChromeOS.

[Siwastaja]

But first you'll have to embrace the idea that paying a corporate entity money and praying they do what you want with it is not the only way to get good software. And then second, you'll have to not belittle people willing to give you their work for free for their own reasons..

Of course, this is all ground we've covered before, and you remain stubbornly resolute in doing nothing unless a company sells you what you want.

Yeah. The actual mechanism for popularity of open source software is neither absence of price tag, nor nerds in their mother's basements doing work for others for free just for getting kicks out of replacing closed source software by their inferior open source counterparts people then use as a "free of cost" low-quality replacements.

The most important mechanism is as follows:
1) You - individual or company need something yourself
2) It does not quite exists
3) So you pick open-source project, or pieces of open source code which help you solve the problem, for yourself
4) You solve your own problem
5) You release your added work, either because you want to (at that point, it's not much extra effort), or because the license forces you to
6) Rinse and repeat, and everyone benefits

This is exactly why corporations want to use open-source software, too.

[Randy222]

IIRC, linux desktops have a market share of about 2% which is hardly an incentive (or business case) for big players to port their apps to linux. Despite this low market share there are companies writing and selling software for linux and many of them are quite successful, i.e. they have paying customers.

Are we now constraining the argument to just "desktop"?
What apps for big players? Apps alone are not an underlying metric, workload-$ is.
Context is needed.

Most desktops don't do much workload-$ at all. The things they connect to do (server backends).

World market share (server OS), 63% Linux, 36% Windows, rest Other.

The OS responsible for workload-$ is predominately NOT Windoze.

[SteveThackery]

Which is fair enough, but when others use those applications and prefer not to switch to the paid-for alternatives (remember the majority of 'apps' which are not proprietary are portable, so you're actually attacking users of every OS), you decide that must be because of their religion and blame them for your preferred apps not being available, rather than the people who actually decide that.

Thank you for this: a properly considered criticism expressed in polite but firm language, with none of the angry tirade to which I've been repeatedly subjected. I appreciate it.

I think you make a good point. I think the issue with Linux desktop apps is a bit like the chicken and the egg: Linux-on-the-desktop won't attract new users with the apps currently available; and potential app developers won't develop apps for such a small user base. I'm not sure how to break out of that loop.

I realise we will differ on this, but I think the current model for producing Linux desktop applications doesn't work very well. In general the apps look klunky and old-fashioned, with poorly thought-through UIs, bumpy workflows, and challenging learning curves. (My opinion.)

I think this is because writing a good application needs more than just coders. Designing the workflow and the UI are highly skilled jobs, whereas in the Linux desktop world it seems like all the work is done purely by coders, with no real involvement of those other skills. I don't know why that is. As you know, I believe that the only way out of this is to point the spending power of the big corporates at the problem, but I admit I don't know how to achieve that. And I embrace your point that it's not just the users expecting software to be free, but also the corporates seeing a tiny user base.

You don't need to be a programmer to contribute. But first you'll have to embrace the idea that paying a corporate entity money and praying they do what you want with it is not the only way to get good software. And then second, you'll have to not belittle people willing to give you their work for free for their own reasons.

Second point first: I don't recall belittling them, but I have described them as geeks, and I think that is probably reasonable. Coding is a singular skill that seems to attract a specific type of person. I was surrounded by them when I worked at BT Labs, so I am familiar with them. I've noticed that, because of their high intelligence, applied with a laser focus, they aren't that good at seeing things from a layperson's perspective. Hence the issue with workflows and UIs: they can't understand what the difficulty is. Of course I am making a sweeping generalisation, and each individual is unique, but I think it is a fair observation. Also, I would challenge your assertion that they give me their work for free. I think they do their work for other geeks, not for people like me. Again, I'm making a generalisation and I might well be wrong.

First point second: I don't give a company money and pray they do what I want. I evaluate their offering - along with that of their competitors - and if it already does what I want, then I give them money. If their product doesn't float my boat, it's tough shit on them - there are always alternatives.

You say that paying money to a software company is not the only way to get good software. Hmmm... I remain to be convinced. A lot depends on how you define "good". If "good" means software that does one, perhaps obscure, thing exceptionally well, without all the overhead of a mass-appeal UI, then yes - a basement geek working for the community can deliver that. If "good" has to include accessibility for a wide range of average punters like me, thus requiring a transparent, logical workflow and an open, attractive and easy UI, then no, I don't think the current model can deliver that. Simply because I see little evidence of it.

Of course, this is all ground we've covered before, and you remain stubbornly resolute in doing nothing unless a company sells you what you want.

OK, as I say, the evidence I have seen suggests I am not wrong. However, I believe you are implying that I shouldn't complain about something unless I'm prepared to be part of the solution. It's a commonly made argument. Let me put a counter argument to you: it is absolutely legitimate to position myself as a consumer only, which is what 99% of computer users around the world do. As a consumer it is reasonable to mention that the applications in this particular ecosystem don't seem very appealing or easy to use, whereas the apps in that ecosystem are generally more appealing. It is also legitimate for said consumer to speculate on why that is.

[SteveThackery]

Are we now constraining the argument to just "desktop"?

Yeah, we have been doing, because the thread is really about Windows 11, and that is resolutely a desktop OS.

Most desktops don't do much workload-$ at all. The things they connect to do (server backends).

World market share (server OS), 63% Linux, 36% Windows, rest Other.

The OS responsible for workload-$ is predominately NOT Windoze.

I wonder if that is true. Certainly in some big companies the desktop PC is little more than a dumb terminal used for kicking off various database queries and reports. But in the corporation I worked, a hell of a lot of work got done locally on the desktops: writing reports in Word, analysing data in Excel, preparing presentations in PowerPoint, drawing complex diagrams in Visio, writing small software tools in Visual Studio. All this was done on the PC itself, albeit the work was stored on managed servers.

And yes, for stuff like project management and financial management, we used server-hosted software.

So, I don't think what you've said is really true; I think PC-based MS Office does a lot of heavy lifting in the corporate world.

[PlainName]

IIRC, linux desktops have a market share of about 2% which is hardly an incentive (or business case) for big players to port their apps to linux. Despite this low market share there are companies writing and selling software for linux and many of them are quite successful, i.e. they have paying customers.

Are we now constraining the argument to just "desktop"?
What apps for big players? Apps alone are not an underlying metric, workload-$ is.
Context is needed.

I don't care (within limits) about desktop workload. I do care about how smooth it is to do stuff, without sharp edges. For instance, I loath those apps where you change some setting and it's saved. No second thoughts - it is done and you're stuffed if it was the wrong thing. Similarly, apps where changes cannot be undone - does wonders for discoverability.

Those are apps and that kind exist on all OSs, but the same thing happens on the desktop. I dislike the MacOS global menu, the single-button mouse (yes, it's gone but it's an example). Windows 10 inability to reverse up a settings thread without making you restart at every level. Do I care if a nice one takes 1% more CPU? Of course not - a crappy system just won't be used so resource usage is irrelevant.

Most desktops don't do much workload-$ at all. The things they connect to do (server backends).

Uh... not everyone is umbilically attached to the cloud. Here, everything is local. I can do everything I normally would (except get to EEVBlog) if my network was unplugged.

But you're talking about apps there, not the desktop. There are things the OS-as-desktop does that is different between systems. The aforementioned global menu being an example. There are utils that can change how the desktop works (the excellent Listary, for example, which I would really hate to be without), but these are not apps per se - they are things that support apps.

World market share (server OS), 63% Linux, 36% Windows, rest Other.

Yes, no doubt. But we are not talking servers. You might as well discuss world market share for chocolate buttons for all that statement is relevant.

The OS responsible for workload-$ is predominately NOT Windoze.

Who cares besides you? The world preference for drinking plain tap water isn't going to make me stop drinking coffee or sparkling bottled and flavoured water in preference.

[SteveThackery]

The most important mechanism is as follows:
1) You - individual or company need something yourself
2) It does not quite exists
3) So you pick open-source project, or pieces of open source code which help you solve the problem, for yourself
4) You solve your own problem
5) You release your added work, either because you want to (at that point, it's not much extra effort), or because the license forces you to
6) Rinse and repeat, and everyone benefits

^ ^ ^ ^ ^ ^ ^
100% THIS

And this is exactly why Linux apps end up as such a patchwork mess, with random bits of functionality bodged on here and there, wherever there's a gap; a menu entry or toolbar icon hastily tacked onto the end of something vaguely related; and no real effort made to make this new feature easy to use because the boss is on at you to get it done and anyway, this is your problem and your solution, so as soon as it's done you're on to the next thing.

Software developed like this, with no proper "ownership" or stewardship of the whole, has no chance of staying smooth, clean, elegant, coherent. For that to happen a software product needs an owner, a steward, who has a compelling interest in the quality and wide appeal of the product, and the authority to direct and control it. Much like everything Adobe produces. Ditto the other big software houses.

Also, exactly like Torvaulds does for Linux. We need a mini-Torvaulds for each Linux desktop application.

[Randy222]

The OS responsible for workload-$ is predominately NOT Windoze.

Who cares besides you? The world preference for drinking plain tap water isn't going to make me stop drinking coffee or sparkling bottled and flavoured water in preference.

workload-$ is how money is made from computing. If you don't care about that and are painting a play field in some far corner, so be it.

99+% of desktop cpu cycles, and IIRC somewhere near 75% of server cpu cycles, are wasting tons of kWh doing nothing.

So in real world, workload-$ does matter to those interested in making money. Way too many $400+ laptops out there doing nothing but reading email and spying on you (probably more cpu cycles spying).

The whole desktop argument (market share, etc) is a bit pointless.

Windoze vs Other arguments need to be fairly constrained to specific realms, like efficient computing, or security, or flexibility, etc etc.

Have I had Windoze crash on me? Yep, lots of times, freezes, problems, etc.
Have I had a nix core dump, yes, but rare instances of installation of app being bad.

A few moths back my upgrade from win10 to win11 seemed to go ok, then all of a sudden it would boot into a "we have a problem" then reboot into "recovery". I could never fix the issue, so I had to re-install Win11 with the "save my sheet" option. Now it's more stable but I don't have confidence in it.

In corp world, end-users don't care if there's an issue, they call IT Helpdesk to get a new machine asap. It's not their stuff, so the extra 4hr coffee break is welcomed. If it were their hardware and their money, then I suspect the tune they fiddle would be a bit different.

Windows is a dominant desktop OS? Surely yes, but end-users only bang keys on what they are given. Give them a Mint desktop with the apps they need, and they'll do the work just the same.

[John B]

And this is exactly why Linux apps end up as such a patchwork mess, with bits of functionality bodged on wherever there's a gap, a menu entry or toolbar icon hastily tacked onto the end of something vaguely related, and no real effort made to make this new feature easy to use because the boss is on at you to get it done and anyway, this is your problem and your solution, so as soon as it's done you're on to the next thing.

What is a "Linux app" though? Most software I use, both major projects and small ones, are all cross platform now. For small dedicated projects, it's hardly a function of the OS whether it's a bodge job or not.

Software developed like this, with no proper "ownership" or stewardship of the whole, has no chance of staying smooth, clean, elegant, coherent. For that to happen a software product needs an owner, a steward, who has a compelling interest in the quality and wide appeal of the product, and the authority to direct and control it. Much like everything Adobe produces. Ditto the other big software houses.

Also, exactly like Torvaulds does for Linux. We need a mini-Torvaulds for each Linux desktop application.

That's a huge generalisation. Even in community based projects, people naturally will form hierarchies.

A good example I like to examine is in the home automation space. Home Assistant is controlled by a company, OpenHAB is a community project. Both maintain their own character and coherent direction, but HA being run by a company hasn't necessarily made it better off in a technical sense, however it does add additional monetary priorities.

As a counterpoint, is Windows coherent now? W11 seems to have a mish-mash of new and legacy settings applications, UI changes and alterations to core legacy programs that were neither needed nor wanted.

[Randy222]

The most important mechanism is as follows:
1) You - individual or company need something yourself
2) It does not quite exists
3) So you pick open-source project, or pieces of open source code which help you solve the problem, for yourself
4) You solve your own problem
5) You release your added work, either because you want to (at that point, it's not much extra effort), or because the license forces you to
6) Rinse and repeat, and everyone benefits

^ ^ ^ ^ ^ ^ ^
100% THIS

And this is exactly why Linux apps end up as such a patchwork mess, with bits of functionality bodged on wherever there's a gap, a menu entry or toolbar icon hastily tacked onto the end of something vaguely related, and no real effort made to make this new feature easy to use because the boss is on at you to get it done and anyway, this is your problem and your solution, so as soon as it's done you're on to the next thing.

Software developed like this, with no proper "ownership" or stewardship of the whole, has no chance of staying smooth, clean, elegant, coherent. For that to happen a software product needs an owner, a steward, who has a compelling interest in the quality and wide appeal of the product, and the authority to direct and control it. Much like everything Adobe produces. Ditto the other big software houses.

Also, exactly like Torvaulds does for Linux. We need a mini-Torvaulds for each Linux desktop application.

Just for clarity sake, not every app compiled down for use on any specific OS, is opensource or patchwork or bodged.
Just as one pays for some CAD or simulation app for Windows, the same is done for Linux.

For something to be good, it has to involve money. So then we get to talk about...... are you happy with what you're paying for?

I have used plenty pay-for apps where there's an installer for Windows and Debian, some work way better on Windoze, some better on the Linux.

Is opensource different than "free"? All opensource is "free", but not all "free" is opensource. Some don't like the free Acrobat Reader, maybe they should make that opensource?

[m k]

albeit the work was stored on managed servers.

Networked disk space.

for stuff like project management and financial management, we used server-hosted software.

Workload-$.

I think PC-based MS Office does a lot of heavy lifting in the corporate world.

Dumb terminal.

[SteveThackery]

What is a "Linux app" though? Most software I use, both major projects and small ones, are all cross platform now. For small dedicated projects, it's hardly a function of the OS whether it's a bodge job or not.

I was talking about open source apps, which almost all Linux desktop apps are but you are right - I should have made that clear.

Actually I'm not such a fan of cross-platform apps because each OS has its own look-and-feel and UI philosophy, and it's hard to observe those when using a cross-platform tool chain. I think some apps do it better than others. SoftMaker Office, for example, looks almost identical on every platform it targets, including Android, and as such looks distinctly odd on an Android tablet because it ignores most of the Android look-and-feel and ways of working.

I felt the same way about the Affinity suite: it was too obviously a MacOS application first, ported to Windows. Therefore the functionality was almost all accessed via menus and keyboard shortcuts, with very little exposed on toolbars - the opposite from typical Windows apps.

Anyway, sorry, this is off-topic, but an interesting aside.

As a counterpoint, is Windows coherent now? W11 seems to have a mish-mash of new and legacy settings applications, UI changes and alterations to core legacy programs that were neither needed nor wanted.

In my opinion Windows went seriously off the rails after Windows 7, and it remains a mess to this day. I think Microsoft made a massive mistake by trying to make Windows all things to all people. They decided to change it to a touch-first UI with the launch of W8. Actually I really liked W8 on my Windows 2-in-1 - the touch UI was exciting and new, with some innovative elements - but it was a car crash for mouse and keyboard users. Then they launched W8.1, which rolled back some of the more out-there features of W8 and proved how little courage Microsoft had in its convictions.

Windows 10 was solid but still had a largely touch-oriented UI* except, that is, for all the W7-style stuff in the background that they hadn't got round to converting. Quite the mess. And personally I find W11 quite underwhelming - a lot of noise for little more than a new Start menu and task bar. Plus hundreds of trivial and largely pointless tweaks.

Microsoft's biggest mistake was to try making Windows work for both touch-oriented users and keyboard/mouse users. They ended up with an enormous mess. Apple got it right: different horses for different courses: macOS for the desktop workstation, iPadOS for the tablets, iOS for the phones, and watchOS for the watches. Of course they are all derived from the same Unix core, but each one is meticulously tuned to its target platform. APIs are made as shared as possible, but with zero leaway where sharing is inappropriate.

Apple got it absolutely spot on, technically, and Microsoft got it wrong: one OS and one UI can never work well across the whole piste. Fluent* is a failure. The expression "jack of all trades, master of none" comes immediately to mind.

Phew! Quite the rant. But that's why I look longingly at Linux Mint - it resembles Windows before it all fell apart.

------

* I'm talking about Fluent. From ChatGPT:

Fluent is Microsoft’s cross-platform UI design system intended to scale across:
Desktop (Windows)
Web
Mobile
Mixed reality (HoloLens)


Here is another excerpt from ChatGPT, summarising the different philosophies:

You can summarise the three philosophies like this:
Apple: “The UI should disappear.”
Google: “The UI should explain itself.”
Microsoft: “The UI should adapt to everything.”               <------- stupid idea (not ChatGPT)

[SteveThackery]

I think PC-based MS Office does a lot of heavy lifting in the corporate world.

Dumb terminal.

No - installed and running on the desktop PCs.

[Bicurico]

These Windows vs. Linux discussions almost never go anywhere because a basic framework is never established beforehand. People end up talking past each other.

First question: what context are we even talking about?

Generic OS vs. desktop OS

Are we talking about operating systems in general (servers, embedded devices, routers, appliances, mobile), or are we specifically talking about desktop PC usage?

I consider Unix/Linux/BSD excellent for servers, embedded systems, and infrastructure. Being able to SSH into a router, run BusyBox on a satellite receiver, or manage a headless system remotely is exactly where these systems shine.

Linux as a desktop OS for my personal workstation is a completely different matter. In fact, my opinion is almost the opposite. So yes, I can simultaneously think Linux is excellent and terrible—depending entirely on the use case.

If this distinction isn’t made upfront, the discussion is already broken.

Open source vs. commercial software

Are we discussing open-source tooling, or professional commercial software?

I have no issue using GNU Radio on Linux or Windows—the experience is essentially the same. The same applies to SDRangel, which is a brilliant open-source project.

But when I need to work, I use software like Autodesk PowerMILL, which only exists on Windows. In my view, it is the best all-purpose CAM software for 3- to 5-axis milling and robot programming. There is nothing remotely comparable on Linux.

So the choice is trivial: I use the OS that runs the software I need.

This is no longer “hacker/nerd” territory where the answer is “just write your own CAM software.” Professional, high-end industrial software is extremely expensive to develop and must be commercially exploited to justify that investment.

Linux fundamentally struggles here. At its core, everything is built on GNU compilers and libraries, which effectively pushes projects into a GNU/open-source ecosystem by default. That severely reduces the incentive to develop true commercial desktop software for Linux. Yes, there are paid Linux products—but they usually monetize services, cloud access, or add-on binaries, not fully proprietary desktop applications in the traditional sense.

Drivers, certification, and liability

Then there is the proprietary driver problem.

As far as I know, Linux still does not offer the same level of GPU driver support as Windows—largely because the best drivers are closed source. In CAD/CAM/CAE environments, this matters enormously.

In aerospace and other regulated industries, certified drivers are mandatory. If you don’t use them, your work can be rejected outright.

Now imagine a finite-element simulation where a GPU driver has a subtle numerical error. That error propagates through a chain of decisions, and eventually something fails catastrophically. That is not a hypothetical risk—it’s exactly why certification exists.

This alone disqualifies Linux desktops from many professional workflows, regardless of ideology.

Which “user” are we talking about?

When people praise the greatness of Linux, which user are they evaluating it for?

A power user / nerd who enjoys debugging systems?
Or a professional who just wants to get the job done and is otherwise not OS-literate?

From a user-experience perspective, most Linux distributions are simply bad. They work fine if you stick to the preinstalled software (browser, office suite). The moment you introduce less common hardware—SDRs, satellite tuners, specialized IO cards—or try to install non-trivial software, everything starts to fall apart.

Wrong kernel. Wrong headers. Missing dependencies. Conflicting package versions.

And suddenly you are completely on your own.

There is no real support chain. The developer of software X depends on library Y, which depends on Z, and nobody knows why Z isn’t available on your system. At that point, you’re not being productive—you’re debugging the OS.

So the real question is: what are we actually debating?

Because “Linux vs. Windows” is meaningless without first agreeing on:

the use case (server vs. desktop),

the software model (open source vs. commercial),

the regulatory environment (certification, liability),

and the type of user involved.

Without that framework, these discussions are just noise.

[soldar]

I can only speak for myself, a home user with only basic needs, internet surfing, email, office suite, Google Earth, Emule, etc.

For me and those basic needs Linux Mint is working fine and better than Windows ever did.

Compared with Windows, Linux Mint

- Requires less resources and will run on HW which will not run Windows.
- Runs faster that Windows
- Has no telemetry or other annoying features of the sort
- No need for any account with MS or anybody
- I can control updates 100%. I choose if and when.
- I am totally in control of my computer

Again, just me. People do not care about OS, they care about apps or programs. If you want a program which only runs on Windows then Windows is what you need. But I do not specifically need any Windows program.

I am running a 15 year old computer which does everything I need at lightning speed. No accounts, no telemetry, no ads, it only does what I need.

The alternative is buying a new computer every few years just to please MS. No thanks.

I install Linux Mint and it just works and does what I need. Most of my friends and family are in the same situation and I have installed Mint on their computers when they would no longer run Windows and they are all happy. They have only basic needs, like myself. For me, as a home user, no expert, I prefer Linux Mint to Windows by a long shot.

My wife has a Windows laptop for work because it is what her employer requires and provides.  When she has problems, and she does have them once in a while, she talks to their IT people and sometimes they resolve it and sometimes they just say they will look into it.

[Zucca]

Yesterday I downgraded my Lab EE PC from W11 to W10 and deactivated the TPM in the BIOS.

All the rest of my computers will run Linux.
I also sold all my Microsoft stocks.

Thanks Microsoft for the ride! Bye bye for now.

[madires]

As a sidenote, RAM prices are insane at the moment (AI bubble). 32 GB DDR5 was about 80 bucks last summer and now it's 400 bucks. Anyone needing a new PC for Win11 will be delighted.

[SiliconWizard]

As a sidenote, RAM prices are insane at the moment (AI bubble). 32 GB DDR5 was about 80 bucks last summer and now it's 400 bucks. Anyone needing a new PC for Win11 will be delighted.

This is absolutely awful. If you use DDR RAM in one of your own products, this is also delightful. Unless you had stock for at least a year or two.

[SteveThackery]

These Windows vs. Linux discussions almost never go anywhere because a basic framework is never established beforehand. People end up talking past each other.

First question: what context are we even talking about?

Generic OS vs. desktop OS

Are we talking about operating systems in general (servers, embedded devices, routers, appliances, mobile), or are we specifically talking about desktop PC usage?

[ENORMOUS SNIP]

Without that framework, these discussions are just noise.

So you haven't read the thread.

The question was "Is Windows 11 dying?" The thread seemed to acknowledge that W11 can only die if there is a viable alternative, otherwise people will stay with Windows. The discussion then made the logical step to debating whether Linux is a viable alternative to Windows.

I don't accept that the things in your list have to be nailed down first. Here is where we are:

 - the use case (server vs. desktop)   We are discussing W11, so desktop

 - the software model (open source vs. commercial)   We've been discussing both

 - the regulatory environment (certification, liability)    Not something W11 users think, care, or know about

 - and the type of user involved   It is generally accepted that W11 is for typical end users, not system admins

[m k]

I think PC-based MS Office does a lot of heavy lifting in the corporate world.

Dumb terminal.

No - installed and running on the desktop PCs.

Document generator loaded from local storage.
Nothing heavy, merely a typewriter of today.

[SteveThackery]

I can only speak for myself, a home user with only basic needs, internet surfing, email, office suite, Google Earth, Emule, etc.

For me and those basic needs Linux Mint is working fine and better than Windows ever did.

With the exception of Office, these are pretty much all content consumption applications. You could use almost any OS for that, including Linux of course.

Personally I wouldn't recommend Windows or Linux for that workload - it is better served by Android or even ChromeOS. If you and your friends want to continue using (relatively) ancient desktop computers, then Linux is a much better option than Windows.

However, I find that an Android tablet is much more convenient for a consumption-based workload of the type you describe. Get rid of those prehistoric desktops and you will liberate a significant and useful amount of space in your homes. 😀👍

[coromonadalix]

you can  subst everything, os  office tools, emails etc ...

but no access to any Msoft 365 services like onedrive, sharepoint  etc ....   

we tried linux,  does not cope well on our business infrastructure  ... credentials  security  deployment  management ...   and lastly business central ...

[PlainName]

but no access to any Msoft 365 services like onedrive, sharepoint  etc ....   

What's the downside?   

[paulca]

but no access to any Msoft 365 services like onedrive, sharepoint  etc ....   

What's the downside?   

Scarepoint sharepants.  But make sure and open a ServiceNever request to access shitepoint.

Similar to confobulence.

Jira-niyoru Pikachu?

[Bicurico]

These Windows vs. Linux discussions almost never go anywhere because a basic framework is never established beforehand. People end up talking past each other.

First question: what context are we even talking about?

Generic OS vs. desktop OS

Are we talking about operating systems in general (servers, embedded devices, routers, appliances, mobile), or are we specifically talking about desktop PC usage?

[ENORMOUS SNIP]

Without that framework, these discussions are just noise.

So you haven't read the thread.

The question was "Is Windows 11 dying?" The thread seemed to acknowledge that W11 can only die if there is a viable alternative, otherwise people will stay with Windows. The discussion then made the logical step to debating whether Linux is a viable alternative to Windows.

I don't accept that the things in your list have to be nailed down first. Here is where we are:

 - the use case (server vs. desktop)   We are discussing W11, so desktop

 - the software model (open source vs. commercial)   We've been discussing both

 - the regulatory environment (certification, liability)    Not something W11 users think, care, or know about

 - and the type of user involved   It is generally accepted that W11 is for typical end users, not system admins

In this case, my answer is: no Windows is not dying.

People may be swapping out HOME USE computers/laptops for mobile devices (smartphones and tablets running Android or iOS), but the vast majority of WORKING people will continue to use Windows for times to come, simply because this is what drives the professional/commercial applications they rely on.

Percentage-wise, a noticeable drop will be noticed, especially if you consider ALL use cases: the drop is against mobile devices.

But "dying" assumes it will disappear, and I am 100% sure it won't disappear in the foreseeable future.