Windows 11 is dying....

[ejeffrey]

No it does not because at the end of the day, you can control what's in there and even wipe it if you so please. So all control remains with the user. As for the OS, the mere fact that you run OS presumes that you trust it's vendor since it by definition has access to all your hardware and can control it. So not trusting OS vendor but still running their OS doesn't make a whole lot of sense to begin with.

Well, part of my trust in an OS is being able to see and control what it does.  And Windows 11 is reducing that further.  For instance, you can't (without unsupported and undocumented hacks that will eventually be blocked) run windows 11 in a hypervisor with a soft TPM provided by the hypervisor. 

It's not yet at the android/ios/ChromeOS level where only signed apps approved by the OS vendor can run, and can run and only with restricted permissions, or if an alternative is allowed it's only in a super restricted sandbox.  But that's definitely the goal, and honestly that's the only thing that can actually provide the security that MS _claims_ that TPMs do.

MS also attempted but ultimately backed off of making trusted boot mandatory (not able to be disabled) for windows supporting systems -- AFAIU it's already this way on ARM but not yet x86.  This makes booting any OS dependent on Microsoft signing the bootloader shim.  And this bootloader shim is a major source of "security vulnerabilities" (really: secure boot bypass bugs), so it's now going to be continually in the crosshairs for increased restrictions.

[asmi]

Well, part of my trust in an OS is being able to see and control what it does.  And Windows 11 is reducing that further.  For instance, you can't (without unsupported and undocumented hacks that will eventually be blocked) run windows 11 in a hypervisor with a soft TPM provided by the hypervisor. 

I haven't tried it myself, but I'm pretty sure you can run Win11 in a VM. That is a major testing workflow used for many different purposes - from debugging device drivers to testing upcoming updates in corporate environments before rolling them out across the company, to testing in-development software in different software and hardware environments.

It's not yet at the android/ios/ChromeOS level where only signed apps approved by the OS vendor can run, and can run and only with restricted permissions, or if an alternative is allowed it's only in a super restricted sandbox.  But that's definitely the goal, and honestly that's the only thing that can actually provide the security that MS _claims_ that TPMs do.

Actually they can not do that because their primary market relies on a crap ton of custom software developed specifically for those companies, and they are not going to share with anybody because it might disclose some corporate secrets. So that can't possibly be their goal. Maybe for home edition they can get away with it, but not for their main cash cow for sure.

MS also attempted but ultimately backed off of making trusted boot mandatory (not able to be disabled) for windows supporting systems -- AFAIU it's already this way on ARM but not yet x86.  This makes booting any OS dependent on Microsoft signing the bootloader shim.  And this bootloader shim is a major source of "security vulnerabilities" (really: secure boot bypass bugs), so it's now going to be continually in the crosshairs for increased restrictions.

You can always add your own key and sign bootloader with it instead of using somebody else's key.

[Marco]

Well, part of my trust in an OS is being able to see and control what it does.  And Windows 11 is reducing that further.  For instance, you can't (without unsupported and undocumented hacks that will eventually be blocked) run windows 11 in a hypervisor with a soft TPM provided by the hypervisor. 

I doubt Microsoft will ever require attestation, millions of VM's would break. Soft TPM's with UEFI firmware with pre-enrolled keys will keep working.

[shapirus]

FWIW I had to modify my win11 installation image, or something along the lines (maybe some special magic during boot?), to disable the secure boot check, in order to install it in virtualbox. It works.
There have been news since that it was no longer possible to disable secure boot, I haven't checked that.

[PlainName]

I can certainly accept the truth of that, but in my experience it was pretty quick and very stable.

How about you install it, connect to the Internet and count seconds until some kind of worm or virus infects it?

You're applying 2026 knowledge and experience to a 1999 product, and then picking the worst possible case to justify your fearmongering.

Back when W2K was around things were very, very different to what they are now. You wouldn't point to Y2K Internet Explorer and then postulate that it's really very crap to browse random websites now. Or maybe you would, because you don't see how silly that comparison is. At least it wasn't an Apple machine, which got infected just by being powered on, seemingly 

[asmi]

You're applying 2026 knowledge and experience to a 1999 product, and then picking the worst possible case to justify your fearmongering.

Back when W2K was around things were very, very different to what they are now. You wouldn't point to Y2K Internet Explorer and then postulate that it's really very crap to browse random websites now. Or maybe you would, because you don't see how silly that comparison is. At least it wasn't an Apple machine, which got infected just by being powered on, seemingly 

Well people are saying that this 1999 year product is better than a modern one, so I offered a way to check if that's the case. Since there were no Win11 in 1999, the only way to check that is in today's environment.

[SteveThackery]

I can certainly accept the truth of that, but in my experience it was pretty quick and very stable.

How about you install it, connect to the Internet and count seconds until some kind of worm or virus infects it?

I guess I'd be waiting a long time behind my router with NAT?

[nctnico]

I can certainly accept the truth of that, but in my experience it was pretty quick and very stable.

How about you install it, connect to the Internet and count seconds until some kind of worm or virus infects it?

I guess I'd be waiting a long time behind my router with NAT?

Indefinitely... Unless you have an internet provider which is foolish enough to encourage their customers to have their PCs connected to internet directly. There should be a law against such practises.

[PlainName]

You're applying 2026 knowledge and experience to a 1999 product, and then picking the worst possible case to justify your fearmongering.

Back when W2K was around things were very, very different to what they are now. You wouldn't point to Y2K Internet Explorer and then postulate that it's really very crap to browse random websites now. Or maybe you would, because you don't see how silly that comparison is. At least it wasn't an Apple machine, which got infected just by being powered on, seemingly 

Well people are saying that this 1999 year product is better than a modern one, so I offered a way to check if that's the case. Since there were no Win11 in 1999, the only way to check that is in today's environment.

If nothing else, it took up a hell of a lot less disk space! The one I have running is <800MB. Until last year, when the PC running it died, I had W2K controlling my mill primarily because the timer interrupt was super-smooth compared to later Windows.

So, for the time, yes, it was pretty damn good and in some respects it is indeed 'better' than W11. Depends on how you want to measure stuff.

[SteveThackery]

Oh, god, I hope that's a joke.

It's not a joke. Most people are resistant to change. So much so that they will believe any BS which supports their resistance due to confirmation bias even if there is no objective basis for it.

I was referring to the suggestion that Microsoft deliberately releases crap versions of Windows alternately, in order to drum up demand for the next version. It's obviously ridiculous, therefore a joke. Right?

[SteveThackery]

Well people are saying that this 1999 year product is better than a modern one, so I offered a way to check if that's the case. Since there were no Win11 in 1999, the only way to check that is in today's environment.

What? Who said that? And anyway, how do you define "better"?

[Cyclotron]

I can certainly accept the truth of that, but in my experience it was pretty quick and very stable.

How about you install it, connect to the Internet and count seconds until some kind of worm or virus infects it?

I guess I'd be waiting a long time behind my router with NAT?

Indefinitely... Unless you have an internet provider which is foolish enough to encourage their customers to have their PCs connected to internet directly. There should be a law against such practises.

You're not wrong.

I think that it's an interesting subject. Prior to NAT, all computers were connected to the Internet. With IPv6, all computers can again be directly addressable on the Internet without NAT.
My ISP provides me with a /56 block of IPv6 addresses for my home use. I still use a firewall, of course, but no NAT is required.
An IPv6 /56 is 2^72 addresses vs all of IPv4 space being 2^32 addresses. The IPv6 /56 offers me 256 /64 blocks, which are the smallest you can properly mask per the standards.  That's still more than enough to directly map all of the IPv4 space into a single block, with plenty of room left over.

What may surprise folks is that most computers today are connected directly to the internet via IPv6 and IPv6-to-IPv4 NAT GWs, and we carry them in our pockets and on our wrists. Carriers manage their mobile networks "differently" from typical ISP networks.
Windows NT and 98 both supported IPv6. Though not at the level of support it is today on hosts. The auto-addressing scheme came later, and there's more than one.

[Electrodynamic]

I love freecad and have been using it for years.

Cool, I have a real use case question for you. So, say, I'm making a fishing lure. My current flow is: use Inkscape to draw the side profile shape, mark the drilling points etc., also manually draw the top/bottom profile shapes, then print them, cut and proceed to the woodworking fun.

Inkscape isn't too great for all this, because it requires a lot of manual operations and approximations, especially when I try to estimate the volume of the model. Its measuring capabilities are all super basic, "because Inkscape is not CAD software".

I kinda want to make my models 3-dimensional to make things easier and more predictable: calculate surface area (3d and a given cross-section), volume, measure linear sizes etc., not to mention having a proper 3d visualization.

So if I move my workflow to freecad and obviously decide to keep it simple in the beginning, i.e., create the 3d shape by "extruding" a side profile shape and then adding some chamfering on the corners, how steep and long of a learning curve may it require to arrive at that, what do you, as a long-time user, think?

I've been thinking of freecad, but never really even tried to run it (nor any other 3d modelling software actually), and now that I'm reading all these horrors about it, I'm somewhat scared.

Whenever I want to design something in Freecad I usually look for video's on the subject. Google freecad fishing lures like this video.

Make your first fishing lure in FreeCAD

I follow along and go through the steps then make a few notes on the tools used and process. Once we understand the basic process then it's much easier to start playing around and modifying things. I think it's much easier and faster to learn by example.

I use Freecad for most basic 3D parts but use Blender whenever there are more complex curves.

[Siwastaja]

I was referring to the suggestion that Microsoft deliberately releases crap versions of Windows alternately, in order to drum up demand for the next version. It's obviously ridiculous, therefore a joke. Right?

It's a similar joke/meme to "this year is the year of desktop linux". It is a joke but a carefully crafted joke enough that some people take it seriously.

Once one gets into the mode of confirmation biases, they can easily find some proof. Obviously with each new version of any software, new bugs get introduced. And, with each new version, some old bugs get fixed. Similarly, new features usually do not work perfectly when they are first introduced, but get better on the next release.

Select suitable set of bugs/features and you can easily see how "good" and "bad" versions alternate, on any software.

[nctnico]

I was referring to the suggestion that Microsoft deliberately releases crap versions of Windows alternately, in order to drum up demand for the next version. It's obviously ridiculous, therefore a joke. Right?

I have a feeling you have not been around very long. If you delve into the history of Windows, you can easely spot an alternating pattern where Windows versions where bad / good. I doubt it is on purpose though. Don't attribute to malice what can be explained by stupidity.

[ebastler]

I was referring to the suggestion that Microsoft deliberately releases crap versions of Windows alternately, in order to drum up demand for the next version. It's obviously ridiculous, therefore a joke. Right?

I have a feeling you have not been around very long. If you delve into the history of Windows, you can easely spot an alternating pattern where Windows versions where bad / good. I doubt it is on purpose though. Don't attribute to malice what can be explained by stupidity.

I listed those Windows versions a few posts above, and made the (tongue-in-cheek) suggestion that the alternating pattern might be somewhat deliberate. That's what prompted Steve's comment.

To clarify: I do not think it is deliberate. But I do think there is a mix of bad Microsoft design decisions and user psychology at play:

When Microsoft has made major changes to Windows (often in the area of new UI "paradigms"), their first attempts have tended to be objectively poor, and they fixed this in the next revision -- with tweaks and sometimes significant back-pedaling. At the same time, users were alienated by the changes in the original release, then to some extent got used to them and accepted them in the next release.

[SteveThackery]

I was referring to the suggestion that Microsoft deliberately releases crap versions of Windows alternately, in order to drum up demand for the next version. It's obviously ridiculous, therefore a joke. Right?

I have a feeling you have not been around very long.

Goodness me, how I wish that were true! 😅

However, as someone who didn't get into Windows until Version 3.1, I accept I am a baby compared with you. I do claim one good decision, though: I switched to NT as soon as NT4 was released, and never went back to the original code base after that. I have used every NT version of Windows since. In fact I am one of the few people on the planet who actually liked W8, but only on a touch-based device.

Having lived with Windows since 3.1, I am intimately familiar with it, and of course I recognise the claim that every alternate version is crap. I think there is something in it, although a lot depends on what your usage criteria are - crap for one person might be just fine for another.

But where I dig my heels in is the claim that this is all part of a big conspiracy by Microsoft. To deliberately release crap software in order to encourage the adoption of the next release. It's beyond ridiculous, and anyone who really believes that probably wears a tin foil hat.

[paulca]

I know it's anecdotal but I started a new job this week.

I have, so far, got two MacBook Pro's a 14" and a 16".  No Windows 11 in sight.  (Two because "Company + Customer" laptops)

[SteveThackery]

When Microsoft has made major changes to Windows (often in the area of new UI "paradigms"), their first attempts have tended to be objectively poor, and they fixed this in the next revision -- with tweaks and sometimes significant back-pedaling. At the same time, users were alienated by the changes in the original release, then to some extent got used to them and accepted them in the next release.

I think this is the most credible explanation by far. In general, it takes a couple of versions - maybe more - for Microsoft to polish their latest marvellous idea.

Windows NT 4.0
Windows 2000
Windows XP
Windows Vista
Windows 7
Windows 8
Windows 10
Windows 11
Windows whatever is next

In terms of this list, there are a couple of things I disagree with. I don't think W2000 should have a line through it. It was very good in its own right and a solid step forward from NT4. XP was a further solid step. So I think there's a run of three non-crap versions in the list: NT4, W2K, XP.

Also - and this is controversial - I think the line through V8 should be dotted. It was very awkward for mouse/keyboard users, but excellent for touch UI users. I loved using it on my Sony Vaio 2-in-1 - it was innovative and modern and suddenly made its competitors look distinctly "ordinary".

Finally, I think V8.1 warrants a place on the list, it being a very significant change from V8. Should there be a line through it? Well, only a dotted one. Again it was very good for touch UI users.

Right now everybody is singing the praises of W10. But not me. I think it might warrant a strikethrough for its awful, bodged together, half-touch half-mouse/keyboard UI. Making every dialog twice (or more) as big as it was in W7 is no use for mouse users, and leaving a load of old UI elements in there, carried forward from the XP days, is no use for touch users.

So no, W10 isn't one of the great releases. In my opinion, the best ever mouse/keyboard UI was in Windows 7, and the best ever touch UI was in Windows 8.

Windows 8.1 was a half-baked merging of W7 and W8, and Windows 10 was a two-thirds-baked merging of W7 and W8.

-----
EDIT: To be fair, I acknowledge that each release improved on security. My comments are aimed at usability.

[PlainName]

I have, so far, got two MacBook Pro's a 14" and a 16".  No Windows 11 in sight.

I have several times been tempted to go Apple to deal with W10+, but it's the cost and walled garden that's the sticking point. And global menu. Thankfully they've moved on from the single-button mouse.

[Siwastaja]

Windows 3.1 and before: small gradual improvements
Windows 95: Large fundamental changes, some hated it, some loved it
Windows 98: Small gradual fixes over Windows 95.

So now is 95 a "bad" and 98 a "good" Windows? Some could say so, but I don't think so (especially if by 95 we mean the fixed version, not the initial release)

(Windows NT, a separate parallel product line. Considered that of better design by pretty much everyone, but not considered a general-purpose option, and does not fit into the linear history timescale as such.)

Then after the many versions of 98, Windows 2000. Large fundamental changes, but because those were already tested on NT, was fairly stable and usable, so generally a real improvement. This is definitely a "good" Windows, most would agree I guess?

Then Windows XP. Small gradual improvements over Windows 2000. Saw some hate for reasons unclear in retrospect, really. Then again wasn't really game-changing either. But this coincides with the proliferation of home computers - I mean the period in history where "all those normies who still didn't have a computer" finally got one, and for them Windows XP is their first OS. So around 2001-2008, just before smartphones, was era in history where home computers were more relevant than ever. And Windows XP also was one of the longest-in-use operating system. People really ended up liking it.

So what, do we have TWO good Windows (2000 and XP) after each other? Wasn't they supposed to alternate?

Then Vista, nearly everyone agrees it was crap, everyone hated it. Brought in some breaking new changes and new fundamental ideas which people did not like. This is clearly a bad Windows, and most would agree.

Then Windows 7, which is undoubtedly an improvement over Vista, but IMHO only because Vista clearly sucked. It's the case of introducing new ideas and failing to make people like them on the first try (Vista), but then being able to polish them on the second try (7). Maybe 7 was a sweet spot to some as it still resembled a classic operating system and not a thin client for MS cloud services, I don't know. I was not able to like it and saw too many problems to use it as a daily driver and migrated completely my desktop use to linux in 2014-2015 or so, when Windows XP had been end-of-support for many years already.

But clearly, consensus is that Windows 7 is a "good" Windows.

And then, finally, for all the rest, it seems no one likes any OS Microsoft manages to produce anymore; people are getting more honest about the motivation, they use Windows to run Windows software because it works well enough. There are occasional strong feelings of dislike to Microsoft's practices even among Windows users but it seems this is now regardless of version, same complaints with 10 and 11.

So really, to summary, I can only see the alternating pattern of "good and bad" XP (good) -> Vista (bad) -> 7 (good) -> 8 (bad), and even that is pretty questionable and depends who you ask. The more you look into either history, or more recent versions, the more the theory of altering good and bad versions falls apart, IMHO.

[SteveThackery]

I have, so far, got two MacBook Pro's a 14" and a 16".  No Windows 11 in sight.

I have several times been tempted to go Apple to deal with W10+, but it's the cost and walled garden that's the sticking point. And global menu. Thankfully they've moved on from the single-button mouse.

Absolutely. Apple got it wrong with the single-button mouse, but stuck to it for years because, as far as Apple is concerned, they never make mistakes and their competitors never do something better.

One area where Windows has wiped the floor with MacOS for years is in windows management. Attaching windows to the sides of the monitor, arranging them in tiles or other layouts.... Apple have only recently started to catch up.

[SteveThackery]

But clearly, consensus is that Windows 7 is a "good" Windows.

For me, yes, absolutely.

I consider W7 to be the very pinnacle of the old WIMP UI paradigm (windows, icons, menus, pointer), after years of fine tuning and polishing. If you like that type of UI (I do), then I don't think W7 has ever been surpassed.

Unfortunately for me, later versions of Windows have varying amounts of touch-based UI folded in, which I very much dislike when using my mouse/keyboard workstation. And much to my dismay, even Linux has started along that path! At least, Mint has.

Now, instead of checkboxes and radio buttons, my favourite Linux has adopted those large blue slider switches. Now look, they were designed for poking with a finger, not clicking with a mouse! I wish Mint would stick to the pure WIMP UI and release a separate version for touch-based devices. That is exactly what Apple did.

[ebastler]

Windows 3.1 and before: small gradual improvements
Windows 95: Large fundamental changes, some hated it, some loved it
Windows 98: Small gradual fixes over Windows 95.

So now is 95 a "bad" and 98 a "good" Windows? Some could say so, but I don't think so (especially if by 95 we mean the fixed version, not the initial release)

Agree; the "every other version is good" (for most users) did not apply to the original, DOS-based Windows versions.

Windows 1 and 2 were clearly work in progress; some might say the same for 3.0. I think there's a consensus that 3.1 was the first solid version that gained massive traction, and 3.11 brought some improvements in networking ("for workgroups"). The successor, Windows 95, was pretty good in my view, and as far as I remember was well-received in the market -- despite the comprehensive technical changes (16/32 bits!) and UI changes. I don't have any recollection of Windows 98: All home users I recall staid on 95, and professional use I was involved with switched from 3.11 straight to NT.

So you might argue that Windows 95 was the one and only instance where Microsoft got major changes to the OS right first time. 

[madires]

Windows 3.1 and before: small gradual improvements
Windows 95: Large fundamental changes, some hated it, some loved it
Windows 98: Small gradual fixes over Windows 95.

Windows Me

Then after the many versions of 98, Windows 2000. Large fundamental changes, but because those were already tested on NT, was fairly stable and usable, so generally a real improvement. This is definitely a "good" Windows, most would agree I guess?

Yep, Windows 2000 was great for business applications, i.e. swift and stable.