Windows 11 is dying....

[SteveThackery]

You can disable updates.

Only temporarily, I think?

But I take your point. To me it's like Windows updates are automatic, and it's up to the user to defer them. Linux updates are entirely controlled by the user. It feels different.

[SteveThackery]

I'm sure I've mentioned the conflation of different types of free before. The value of open, free standards and tooling (and yes, that includes desktop apps) cannot be overstated. I value open formats anyone can work with, not just those who've purchased the latest edition of $INSERT_APP_HERE, far more than I value a flashy new UI.

You don't need to mention it - we understand it. I don't believe anyone likes proprietary file formats apart from the vendors themselves. That's why the Open Document Format got established.  Plus the hundreds of other formats that are also effectively open.

If by "tooling" you mean the apps that work with those files, like LibreOffice, then I disagree profoundly. I've said it before: without the motivation to make money, the app development gets left to volunteers, and that's why LibreOffice feels like a '90s product with bits of functionality bodged on and a mess of a UI.

Why should the tools themselves be free and open source? It removes any incentive for a company to invest and innovate, because their intellectual property is given away immediately. I think open source has failed - it results in mediocre software that remains essentially unchanged for years or decades. Also, why should someone's hard work not be remunerated?

So, we agree that data should be stored in open formats because it actually encourages competition - nobody is excluded from trying to develop a superior word processor, for example. But having to open source everything you develop discourages competition. If there's no money to be made, why bother?

[Monkeh]

Why should the tools themselves be free and open source?

Why should they not? An open option does not mean you can't make proprietary software, and it does not imply that money cannot be made.

I think open source has failed

The success of open source development is behind almost every service and piece of software you use every day, including proprietary ones. Even if you want to retroactively limit that sentence to 'for desktop application development', you must consider that said open source desktop software, hardly having failed, is in routine use by those developers and many users, and actively encourages proprietary tooling to make itself worthwhile.

But having to open source everything you develop discourages competition. If there's no money to be made, why bother?

You don't have to open source everything you develop. Nobody is arguing you should. You're strawmanning desperately.

There is plenty of money to be made, for those who consider that the only reason to do anything - and there are reasons beyond money.

E: Self-moderated, I know better.

[peter-h]

You can disable updates for ever.

There are several configs for this, with the most flexible in the registry, or in the admin policy files (editable with .msc stuff). You can google it. Honestly, it is no issue and if you want a PC to stay the same, you can just do that. I always disable auto updates because auto updates can break stuff.

[SteveThackery]

Honestly, I'm not straw manning deliberately. I absolutely accept that I have only a superficial understanding of the ins and outs of open source.

I wonder if I could ask some questions to clarify things, because I am genuinely here to learn, and change my mind if necessary.

Referring to open source applications:

Why should they not? An open option does not mean you can't make proprietary software, and it does not imply that money cannot be made.

So how does that work in practice? Imagine I come up with a great new concept for working on ODF files; I recruit 20 programmers and we spend a year developing this new application. I've had to borrow money to pay the programmers, but the investor was OK about it because he was expecting to see a lot of sales, which will give a great return on the investment.

If I open source this software, others can pick it up, change the splash screen, and put it out there free of charge. They can do that because they didn’t have to pay anyone to write the software. So when it's out there free of charge, why would anyone pay for it? How does the original investor get his money back?

You don't have to open source everything you develop. Nobody is arguing you should. You're strawmanning desperately.

No, not desperately. Not even deliberately.

There is plenty of money to be made, for those who consider that the only reason to do anything - and there are reasons beyond money.

I've got all the usual open source applications on my Mint installation, including LibreOffice. I didn't pay for them. I don't pay to use them, so nobody is making any money from me. So how is the "plenty of money" actually made?

The success of open source development is behind almost every service and piece of software you use every day, including proprietary ones. Even if you want to retroactively limit that sentence to 'for desktop application development', you must consider that said open source desktop software, hardly having failed, is in routine use by those developers and many users, and actively encourages proprietary tooling to make itself worthwhile.

I'm sorry, I was actually thinking of Linux desktop programs. I can imagine that all those Linux servers in the cloud have all sorts of specialised programs to provide the functionality and management tools they require. And I can see the end users paying money to the software developers for maintenance contracts.

Where I'm stuck is when it comes to people like me. Even if LibreOffice (say) is widely used by companies and individuals, they haven't paid for the privilege, and certainly they won't be paying for a support contract because LibreOffice doesn't need support: everyone knows how to use it.

These are genuine questions, not challenges. I have no interest in arguing, except to repeat one last point. When I say "open source doesn't work" it's because I look at the big applications in the Linux desktop world and - in my judgement - they really are old-fashioned, ugly, klunky, with steep learning curves and obscure workflows. It seems to me they don't get the level of attention from their developers that Windows apps do.

[Monkeh]

Referring to open source applications:

Why should they not? An open option does not mean you can't make proprietary software, and it does not imply that money cannot be made.

So how does that work in practice? Imagine I come up with a great new concept for working on ODF files; I recruit 20 programmers and we spend a year developing this new application. I've had to borrow money to pay the programmers, but the investor was OK about it because he was expecting to see a lot of sales, which will give a great return on the investment.

If I open source this software, others can pick it up, change the splash screen, and put it out there free of charge. They can do that because they didn’t have to pay anyone to write the software. So when it's out there free of charge, why would anyone pay for it? How does the original investor get his money back?

Why did you decide to do it as open source? Again, proprietary software can exist. You're conflating my desire for open options with a desire for all software to be open source. Your fantastic new concept can stay proprietary as long as you like.

I and many others believe open source tooling must exist in order to support open standards. That doesn't place any burden or you or anyone else to create it - those who want it will do so and find ways to finance it.

Where I'm stuck is when it comes to people like me. Even if LibreOffice (say) is widely used by companies and individuals, they haven't paid for the privilege, and certainly they won't be paying for a support contract because LibreOffice doesn't need support: everyone knows how to use it.

Who says they haven't? The Document Foundation, the non-profit which supports LibreOffice, took in just about €1.4bnm in 2024. People who use this software, like it, and feel it should continue to exist and develop.. contribute. Financially and otherwise. Yeah, it's a non-profit - as I said, motivations other than money exist. On the LibreOffice site, you'll find they link to individuals and businesses which offer training (after all, nobody pops out of the womb knowing how to drive an office suite, any office suite), targeted development, accelerated bug fixing, and so forth.

[peter-h]

Most mfgs get around the OS requirements.

For example Nokia used some OS stuff, so they published that, but the published bit was of no use to anybody for recreating a phone.

If you develop a product with some OS software in it, you can do that, or you just point people to where it is and the licensing terms (in the user manual). Nobody will be able to rip off your product because they won't get the part which you wrote.

We had a long thread on this recently.

[SteveThackery]

Why did you decide to do it as open source? Again, proprietary software can exist. You're conflating my desire for open options with a desire for all software to be open source. Your fantastic new concept can stay proprietary as long as you like.

Ah, OK, I see. For example SoftMaker Office is available for Linux desktops, but you have to pay for it. Therefore I'm guessing it isn't open source. However, it sits alongside LibreOffice, which is.

I and many others believe open source tooling must exist in order to support open standards.

I don't think I quite see the logic there. Microsoft Word will save documents in ODF and it isn't open source. I don't see why open source standards require open source applications.

The Document Foundation, the non-profit which supports LibreOffice, took in just about €1.4bn in 2024. People who use this software, like it, and feel it should continue to exist and develop.. contribute. Financially and otherwise.

That's an impressive amount of money, and I guess principally from donations. Plus, as you say, payment for accelerated bug fixes, extra features, etc.

This has been very helpful - thank you. I am still left with the same problem I started with, which is that I don't think the applications available to Linux desktop users are, in general, anywhere near as good as those for Windows.

------
PS:  I must say, for €1.4bn I would have expected to see a lot more evolution in the LibreOffice applications. €1.4bn buys a massive amount of manpower, and yet LibreOffice in 2025 looks awfully like it did in 2024 and 2023. If we assume it costs $200k per developer per annum (which is surely generous) then that's 7000 developers. Of course, a lot of the money goes in building costs and the like. Even if half of it goes that way, we're still left with 3500 developers. OK, OK, let's say just one seventh of the total revenue gets spent on developing LibreOffice. 1000 developers. That's a helluva development team.

By way of contrast, Serif in the UK employed approximately 90 employees (so fewer than that actual developers). They released the first two apps after six years, and the third application after 10 years. So 900 man-years (a lot less really) to go from zero to three world-leading products which attracted universal praise.

So what on earth is the Document Foundation spending all that money on?

[Monkeh]

I don't think I quite see the logic there. Microsoft Word will save documents in ODF and it isn't open source. I don't see why open source standards require open source applications.

Because it is about the principle of not being tied to a proprietary tool you do not control. Open standards without open tooling cannot be freely worked with, you remain beholden to others to interact with them. The moment that tooling is inadequate or broken, you are lost - open tooling is a necessity or you will have to reinvent the wheel day in, day out.

Microsoft did not implement support for ODF willingly. They had to be forced into it.

PS:  I must say, for €1.4bn I would have expected to see a lot more evolution in the LibreOffice applications.

My apologies, I made a gross error while multitasking, that was €1.4m. Which certainly makes more sense. In my defense, they're European - they get , and . confused to the detriment of all.

[Cyclotron]

Microsoft did not implement support for ODF willingly. They had to be forced into it.

I don't think that's accurate. Sun created the XML format on which ODF is based for OpenOffice. That work was done by SGML Open and was later renamed the
Microsoft was a founding member and funded OASIS and the development of ODF while also developing its own XML format.

Your statement appears to imply that some legal entity forced Microsoft, and that they supported it unwillingly. Microsoft went to market with what they felt was superior. The ODF became relevant to their products due to market changes. Large customers would not accept products that didn't include it, so MS added the format to their product.  Market forces, or voluntary changes to adapt to changing markets and win business, sound a lot different than what you implied.

That was back when Munich famously decided to move to Open and away from MS. That whole debacle failed and cost a fortune, only to go back to MS a few years later. These costly mass migrations are because MS is bad, the US is bad, something else is bad, but reliance on fuel and power from obviously problematic areas is surreal. There's more anti-MS migrations going on now, and if history repeats as it often does, they'll just go back in and pay huge sums to do so.

[Monkeh]

I don't think that's accurate.

Large customers would not accept products that didn't include it, so MS added the format to their product.

Forced, coerced, incentivised - pick your term. It wasn't their intention to support the format.

[Cyclotron]

I don't think that's accurate.

Large customers would not accept products that didn't include it, so MS added the format to their product.

Forced, coerced, incentivised - pick your term. It wasn't their intention to support the format.

If you feel better by mischaracterizing it, ok. As long as you're just as amicable to characterize Linux's support for any Microsoft protocol or standard as "Linux did not implement support for MS protocol/standard willingly. They had to be forced into it."

[Randy222]

New just in, Win11 just died. RIP.

Win12 was suppose to be out in 2024. Still waiting.

[peter-h]

Not really. Google on win11 versions and update support dates from M$.

"Support" means nothing unless you are a corporate user and need to tick the "supported" box.

Munich town council dropping linux and going back to windoze is unsurprising, if you want compatibility with other people.

A lot of people had a laugh at the time
https://www.itprotoday.com/linux-os/munich-officially-drops-windows-adopts-linux

[SteveThackery]

Win12 was suppose to be out in 2024. Still waiting.

I've often thought how arbitrary version numbering can be. The Windows internal version numbers have differed from the market version number, and some "major" releases have been considered a "point" release internally.



Windows appears to be under continuous development, so deciding when to launch the "next" version - and what will be in it - seems fairly arbitrary.

[madires]

That was back when Munich famously decided to move to Open and away from MS. That whole debacle failed and cost a fortune, only to go back to MS a few years later.

They returned to MS because of a new mayor and successful lobbying by MS. But this is not the end of the story. Schleswig-Holstein (state) goes Linux (25k PCs) and Frankfurt/Main too (13k PCs and 2.5k servers), for example.

[Zero999]

Win12 was suppose to be out in 2024. Still waiting.

I've often thought how arbitrary version numbering can be. The Windows internal version numbers have differed from the market version number, and some "major" releases have been considered a "point" release internally.

(Attachment Link)

Windows appears to be under continuous development, so deciding when to launch the "next" version - and what will be in it - seems fairly arbitrary.

Remember NT is just the kernel, not the entire OS, the same as Linux is a kernel, which is distributed with other software to form a complete OS i.e. a GNU/Linux distribution.

Microsoft clearly change the rest of their operating system more than the kernel. Cynics such as myself, believe they change the UI more than anything else, for marketing purposes.

[paulca]

Consider that the most modern .Net platform is completely cross platform and at least partially open source.

Microsoft had no choice.  To be "Cloud Native" they need to have software which executes on Linux for containerisation platforms.  Without a cross platform enterprise runtime they would have disappeared in the cloud space entirely.

Azure runs primarily on Linux.  Very likely based on "Open Stack" at the core.

[Infraviolet]

Randy222, I'm pretty sure there was never supposed to even be a Wn 11, let alone Win 12. M$ did say that Windows 10 would be the "last" Windows OS and from then on it would just get "service pack"-like "upgrades" rather than fresh new systems.

peter-h, "Munich town council dropping linux and going back to windoze is unsurprising, if you want compatibility with other people." For the most part anyone on Linux can share files with Windows users, and vice-versa just fine. Sometimes formatting gets messed up a bit when docs get transferred back and forth between LibreOffice and M$ Office, regardless of whether the docs change format or whether they stays consistently in docx or in odt format. The same sort of formatting f*** ups happen when transferring a doc back and forth between users of different M$ office versions too. Only a moronic bureaucracy with an absolute obsession with formatting, and no willingness to accept a simpler formatting (just stick to text and images alone), is going to find problems here. Maybe they went back to Windows for other reasons, but if format compatibility with M$ users was their reason then they should have redesigned their document styling to be less likely to break when converted. The trick for any organisation moving from Windows to Linux is a) to keep some Windows computers for any specialists using Windows only software (particular CAD packages or such) until you can pay someone to improve Wine such that they can then run on Linux (hint: any big organisation paying for Wine developers is going to be able to get a good reputation among the geek community) and b) if you're the kind of organisation which hires a whole department to check the size of margins and formatting bars with a ruler, sack them all, embrace some tolerance for variation in this sort of thing.

Cyclotron, "Linux did not implement support for MS protocol/standard willingly. They had to be forced into it." This actually shows up a major difference between M$'s philosophy and Linux. Linux software often does try, though not always achieve, compatibility with Windows file formats, because they know this will make the software more useful to people. M$ prefers to simply ignore the competition, user experience be damned. Linux tries to support common standards because they know than not everyone can be a purist. Windows occasionally tries to sound sympathetic to common standards, but would very much like to have everyone back in to a walled garden.

[TizianoHV]

Sometimes formatting gets messed up a bit when docs get transferred back and forth between LibreOffice and M$ Office

I'm sorry for the developers but I think LibreOffice is one big enemy of Linux.
Most users are fine with a browser and an office suite.
When you download a .docx or .pptx you want to open it and just work. If you find your document corrupted or messed up the day of the exam you won't be happy.
LibreOffice is the bad stereotype of open source. It is "sold" as 98% perfect, better than office, but in reality is slow and fails even with basic stuff. (I think the gui is not an issue, it is a question of getting used).

It is the default suite in most distros because they can trust it in being constant.

I think it shouldn't be the default, they should let you choose when first installing linux since there are better alternatives (even if not free).

[peter-h]

Sometimes formatting gets messed up a bit when docs get transferred back and forth between LibreOffice and M$ Office

That makes Libre pretty much completely useless - yet it was this (the cost of M$ Office) which was the main driver towards linux.

If you are going to pay for a linux office suite, then why bother changing away from windoze? There is frankly no point whatsoever, once you cost in the cost of losing the system support expertise etc.

It was worked out decades ago that the transition from mainframes/minis (and dumb terminals) to a PC on every desk did not save money, largely because every office ended up with an "expert" who wasted a lot of his time fixing people's PCs. Certainly productivity was generally higher with the dumb terminal setup. I used to run an accounting / order processing package in the 1980s, called Anagram, with the "mainframe" being a PC running DR Concurrent DOS. It worked great.

And with linux you probably won't have any local "expert" because almost nobody uses linux - outside the academic and electronics sphere.

[TizianoHV]

If you are going to pay for a linux office suite, then why bother changing away from windoze? There is frankly no point whatsoever, once you cost in the cost of losing the system support expertise etc.

Well, you can use libreOffice (and almost all office suites) on windows. If I/we are moving to linux it is not to be cheap, but to have something better, because we are tired of microsoft deciding what is better for us.

Now I'm using onlyoffice, free when installed on your PC, and seems fine for everyday stuff.

[soldar]

I'm sorry for the developers but I think LibreOffice is one big enemy of Linux.
Most users are fine with a browser and an office suite.
When you download a .docx or .pptx you want to open it and just work. If you find your document corrupted or messed up the day of the exam you won't be happy.
LibreOffice is the bad stereotype of open source. It is "sold" as 98% perfect, better than office, but in reality is slow and fails even with basic stuff. (I think the gui is not an issue, it is a question of getting used).

It is the default suite in most distros because they can trust it in being constant.

I think it shouldn't be the default, they should let you choose when first installing linux since there are better alternatives (even if not free).

I am using Linux Mint and I can generally get by with LibreOffice Writer but sometimes I just go back to Word 2003 because Writer can be a pain.  For example, in Word if I want to replace two consecutive paragraph marks with a single one I use "^p" to represent a paragraph mark and so I replace "^p^p" with "^p" throughout the document and it just works.

Doing the same thing in Writer is a PITA because it uses "$" to search for paragraph marks but only if the paragraphs are not empty. An empty paragraph is "^$" but "$^$" will not find a regular paragraph followed by an empty paragraph.

And then if you want to replace something with a paragraph you don't use "$" but something else. So a paragraph mark is represented by al least three different things depending on context. And for all this you need to "enable regular expressions" so it feels more like advanced programing rather than just editing a document. Forget it.

The people who make these things should just copy Windows programs and stop pretending Linux users are advanced programmers. I am most definitely not.

[peter-h]

If I/we are moving to linux it is not to be cheap, but to have something better, because we are tired of microsoft deciding what is better for us.

But... is it better?

The reality is that M$ Orifice does work just fine. Even Office 1997 (my favourite £10 Ebay job) works perfectly Or Office 2007 if you want to go really modern.

IMHO ultimately the issue is that if M$ "drop support" for Windows XX then a "compliant corporate" cannot run it anymore because they feel they cannot tick the due diligence box. Does linux solve this? It undoubtedly slows it down, but recently I had to install latest Debian on a machine (for somebody else to run code on) and it needed a fairly recent CPU. I actually ran it in a VMWARE VM but it had to be a recent edition.

I guess the most powerful person in a big company is the LGBTQ+ compliance officer. The second most powerful is the windows support compliance officer

The reason Ebay is full of Lenovo mini PCs, ex win10 and not win11 compliant (no TPM module but have UEFI and secure boot) is because a load of big firms chucked them out, wasting many millions. Then "criminal gangs" sell them as "win11" with a Rufus-hacked win11 install

[TizianoHV]

But... is it better?
The reality is that M$ Orifice...

We are talking of Windows 11 / Linux. Not only microsoft office Vs libreoffice.

M$ Orifice does work just fine. Even Office 1997 works perfectly

I would say "in particular office 97". We have a 25 year old laptop with office 97 and I love it. Super fast and solid (everything is so "mechanical", in comparison office 365 is so "disconnected"), too bad for no docx support. 2007 is the perfect middle point: good support and stable, some says it runs with wine.

If you want to go that route, just use windows 7, office 2007 (If I had the choice, I would do that and bin Linux, sorry ). But I wanted to try something different (since you'll need some tweaking to get win7 working, while hoping in not getting malware).