Some folks have had success running it on certain server hardware (Usually HPE Proliant). There are no graphics drivers for x86, so it is X forwarding only.
I would like to see ReactOS succeed for various reasons, mainly philosophical. On the other hand, for practical real-world use cases, it has to compete with several alternative solutions:
1. Just use Windows 11. Yes, it sucks and MS occasionally breaks stuff - but at least hardware and software vendors will develop their code against Win 11 and test it. In other words, you have the highest likelihood that your computer will work as expected with contemporary Windows applications and drivers.
2. Use an older version of Windows. If you want to use old hardware or software, odds are you will get the best experience with whatever version of Windows they were developed/tested against. You have to accept the lack of support for modern software, and you will need to take appropriate security measures such as not connecting it to the internet - but at the same time, it's unlikely that your Windows 98 retro gaming rig is your only computer, so that's probably an acceptable tradeoff.
3. Run WINE on top of Linux (or some other mature open source operating system). This might not be a good solution for the average person, but ticks the box for people who feel strongly pro-open source, or anti-Microsoft. Since Windows compatibility is dictated by Windows' libraries and frameworks and not the kernel, compatibility is likely to be comparable to ReactOS.
I am not saying that this covers every possible use case for ReactOS, but I would posit it covers enough that the majority of people who might contribute or invest into ReactOS will instead pick one of the above options and invest their time and energy elsewhere.
IIRC ReactOS uses and contributes heavily to WINE. So in many ways your #3 isn't far from using ReactOS, and if done correctly it'll be friendlier for the average person than Linux itself.
No, the Wine developers refuse to accept contributions from ReactOS developers or even people who have seen ReactOS code[0]. So any improvements go one way only.
Of course not. You would be surprised how many developers don't even consider using an LLM in their workflow, myself included. Can't wait for this hype to end.
Firstly, neither OpenAI nor Anthropic is profitable, by a wide margin — investors are going to get impatient at some point.
Secondly, people that aren't enthusiastic about this whole thing are already experiencing something of an AI fatigue with all the AI features violently shoved into them by most software products they use. Being involuntarily subjected to slop in various online spaces can't be good either.
Thirdly, remember NFTs? So many people swore they were The Future™... until they weren't. But at least in that case it was much more obvious how stupid the whole idea is. The scale of the hype was also several orders of magnitude less.
Even if all major provides close down, it doesn't remove what's already out there. Glm / minimax / deepseek / gpt-oss may not be at the same level as current frontier, but you can download them and they're still very capable.
Crazy stupid ideas like cars with only touchscreens have still taken a decade to come in and then to get considered ill-advised even though anyone driving a car could tell how bad of an interface it is. We are still not fully out through the other side.
So while OpenAI or Anthropic are maybe not profitable today, they've got at least 5 years to figure it out. And there is already talk of inserting ads into the "chat", but hopefully that does not work!
But really, LLMs are useful (yes, sometimes only in appearance, but sometimes for real), and with that, there will continue to be investment into them until they are made profitable.
I'm saying nothing, i posted the link of the Wine developers claim for why not accepting contributions by ReactOS developers since the post i replied to wrote that ReactOS contributed to Wine.
I believe the integrity of ReactOS's clean room reverse engineering has been called into question in the past when it was found that there were some header or code files with sections that matched leaked Windows Server 2003 code or something like that. Can't recall for sure though.
"In January 2006, concerns grew about contributors having access to leaked Windows source code and possibly using this leaked source code in their contributions. In response, Steven Edwards strengthened the project’s intellectual property policy and the project made the difficult decision to audit the existing source code and temporarily freeze contributions."
The allegations have been taken seriously and since then the procedure for accepting contributions include measures to prevent such further events from occurring. If you or anyone else happen to have any plausible suspicion, then please report it to the ReactOS team, otherwise keeping alive this kind of vague and uncertain connection between some Windows code leakage and ReactOS fits the very definition of FUD: https://en.wikipedia.org/wiki/Fear,_uncertainty,_and_doubt Please stop.
>it'll be friendlier for the average person than Linux itself.
I think the myth that Windows is easier needs to die. The builds targeted at Windows users are very easy to use; You would likely go into the Command Prompt as much as you would with Windows, and the "average person" spends more time on their non-windows phone than they do in Windows.
I am a 30+ years Windows developer, who thought he would never move, but who migrated literally a week ago, the migration was surprisingly painless and the new system feels much more friendly, and surprisingly, more stable. I wrote it up on my blog, and was going to follow it up with another post about all the annoyances in my first full week, but they were so petty I didn't bother.
You are still in the honeymoon phase. I see a lot of those blogpost in the last months.
In a few weeks you will bump into something that isn't simple and friendly and you will curse that stupid linux. Something that trivially works in windows and is impossible or insanely hard in linux. That is often the time people go back. Old habits die hard.
But still you are 100% right. Windows is not easier. I know because I went from dos to linux and only occasionally dabbled in windows. And I have exactly the same sort of trouble as soon as I try to do something non trivial in windows. Including bumping into stuff that should be trivial but suddenly is impossible or insanely hard.
For years I have seen people say that windows is easier, while actually windows is just more familiar.
My (completely non computer savvy) parents and in-laws are on ubuntu/mint since 2009 and it was the best decision ever to switch them over. And they don't understand why people say linux is hard either (though my father in law still calls it 'Ubantu Linox' for some reason :-P )
At the start I had a small doubt if I should push them to macOS (OSX at the time) as then apple's fanatical dedication to userfriendlyness paid off. But I decided against it because I didn't feel like paying apple prices for my own hardware and it seemed ill advised to manage their systems while not using it myself. I'm very glad about that because apple has gone downhill immensely since ~2009 (imo)
I agree. One can just install Linux Mint or Fedora or anything and then Linux is just as friendly to use. You got a desktop, you can use your mouse to start up the browser, install applications with a mouse click, and so forth. You could do without opening up the terminal. Functionally the same as using Windows.
My parents, and my wife's parents, have been doing just that without any trouble whatsoever.
Just browse to netflix.com and log in. Not any different than in windows.
My parents use mail, firefox and libreoffice writer. That's about all they need and it works fine and is way more stable and hasslefree on linux than when it was when they were still using windows (admittedly quite long ago).
And if you are talking about seeing people install the OS, people can't do that for windows either.
It certainly is, because I still don't see GNU/Linux desktops on sale, other than the short lived netbooks movement.
So normal people have stores with other people that they can talk to when they have problems, or just drag their computer into the store.
With Linux it is always the relative that happens to be around, or drive in on purpose, and had to manually install the <insert favourite distro> of the day.
> With Linux it is always the relative that happens to be around
That's certainly true. And it's a chicken-vs-egg problem that's hard to solve. But it doesn't really have anything to do with which system is easier to use. It has much more to do with Microsoft's past unfair business practices (asking shops more for windows licenses if they happened to sell computers with something else than windows on it comes to mind) and the slowness of retail in adapting. Selling computers is way down (most people don't need more than a tablet/phone), selling in physical stores is way down (has moved online). Shops are not going to spend money on training their salespeople in linux. Most of the time they won't even really know windows.
And of course retail GNU/Linux machines that cost 1/4 of a cheap Apple Mac and yet have outsold them by revenue not number of units for nearly a decade now:
Yes this is absolutely happening. This is a real international market with sales in the hundreds of millions of units. This is not some tiny obscure niche that can be skipped over.
You are so eager to reply that you haven't even read the whole comment.
> So normal people have stores with other people that they can talk to when they have problems, or just drag their computer into the store.
Which of those online stores have a physical address for the normal people to do as per my comment?
Linux forums have enough complaints about those fairly prominent Linux-only vendors, even though they are suppose to control the whole stack.
And they also fall into each having their own <favourite distro>, the other part of the comment that you missed as well.
Normal people aren't using SteamDecks for their daily computing activities.
I use Linux in various forms since 1995, and yet I am tired from trying out such alternatives, the only things that makes me consider it again is breaking the dependency on US tech, and even that isn't really happening, given how much from Linux contributions are on the pockets from US Big Tech.
> You are so eager to reply that you haven't even read the whole comment.
Of course I did. I didn't address your objections because I think they don't hold up, that is why.
> Which of those online stores have a physical address for the normal people to do as per my comment?
Leaving out Apple as computers are not its primary product line any more... that leaves Lenovo, the biggest PC vendor in the world, followed by HP, Dell, Asus, Acer.
Only Apple has retail shops worldwide. I do not know of physical stores for any of the others. Maybe some did once, years ago, but that stuff is fading away and dying now. It's all going online.
You can certainly buy Chromebooks in physical stores. Do they fix them? Only warranty repairs, but the point of Chromebooks is that you don't keep your stuff on them, and you don't upgrade them. Rightly or wrongly (that is, mostly wrongly) they are disposable tech.
It is perfectly possible to buy a computer with Linux on it: a choice of Linuxes, from a choice of vendors, in almost any country. No you can't walk into a shop and try it, but you mostly can't from any vendor. Online sales are the default for many things now. No you can't walk into the vendor's shop and get it fixed, but you can't for any of global PC brands either.
If you want that, go to a local small business. If you want Linux, go to a local small business. Same thing.
Sure there are different flavours and distros. That is _not_ a weakness of Linux. Choice is a good thing, even if sometimes it is scary. You can choose your toothpaste and your clothes and your car as well. We manage.
> I wrote it up on my blog, and was going to follow it up with another post about all the annoyances in my first full week, but they were so petty I didn't bother.
This isn't really my arena, but I did happen to recently compare the implementation of ReactOS's RTL (Run Time Library) path routines [0] with Wine's implementation [1].
ReactOS covers a lot more of the Windows API than Wine does (3x the line count and defines a lot more routines like 'RtlDoesFileExists_UstrEx'). Now, this is not supposed to be a public API and should only be used by Windows internally, as I understand it.
But it is an example of where ReactOS covers a lot more API than Wine does or probably ever will, by design. To whom (if anyone) this matters, I'm not sure.
That's an interesting data point. I wonder if there is a hard technical reason why that logic could not be added to WINE, or if the WINE maintainers made a decision not to implement similar functionality.
There is not a hard technical reason, just different goals. WINE is a compatibility layer to run Windows apps, and thus most improvements end up fixing an issue with a particular Windows application. It turns out that most Windows applications are somewhat well-behaved and restrict themselves to calling public win32 APIs and public DLL functions, so implementing 100% coverage of internal APIs wouldn't accomplish much beyond exposing the project to accusations of copyright infringement.
IIRC, there is also US court precedent (maybe Sony v. Connectix?) that protects the practice of reverse-engineering external hardware/software systems that programs use in order to facilitate compatibility. WINE risks losing this protection if they stray outside of APIs known to be used (or are otherwise required) by applications.
There's also another partial Win32 reimplementation in retrowin32, with the different goal of being a Windows emulator for the web, not for Linux or as alternate OS, at https://evmar.github.io/retrowin32/ It thus has an even more sparse path/fileapi.h implementation [2] than WINE and ReactOS. Written in Rust.
That's (part of) my point. A project like ReactOS which clones Windows down to the kernel level solves for a very small set of practical use cases which are not covered by real Windows, or Linux+WINE.
It's worth noting that 30 years ago, there was a definite advantage to an open source operating system which could reuse proprietary Windows drivers - even Linux had a mechanism for using Windows drivers for certain types of hardware. Nowadays, Linux provides excellent support for modern PC hardware with little to no tinkering required in most cases. I have seen many cases where Linux provided full support out-of-the-box for a computer, whereas Windows required drivers to be downloaded and installed.
I think using WINE over Linux has won as the option to consider if you want to run Windows applications on a non-Windows OS without loading Windows into a VM.
There is a growing list of software that which has discontinued support for Windows 10 on the latest versions (or the Server versions thereof). I'm not sure if your example of running a ~16 year old version of SQL Server on Server 2016 demonstrates.
To my original post - if you only need to run an old version of a software package, then an old version of Windows is fine. Just because something is old, it doesn't mean that it is not useful.
The system runs only one app and does not serve public internet content - it does not get any updates at all, only this one app is updated every few month.
Sigh, I hate to agree with you. On a slight tangent, I was exploring what file system I could use safely with different OSes, so that I could keep my personal data on it and access (or add to it) from other OSes, and incredibly NTFS is the only feature rich cross-platform filesystem that works reliably on all the major OSes! None of the open source solutions - ZFS, Btrfs, Ext etc. work reliably on other OSes (many solutions to make them cross-platform or still in beta, for years now). It's the Windows effect - open source developers are putting so much effort into supporting windows tech because of it's popularity, that unknowingly they are also helping it make even more entrenched, to the detriment of better open source solutions.
Last time I looked at this, I think I determined that exFAT also had reasonable support for Windows, Linux, and MacOS? I guess it might not be "feature rich", but it's at least suitable for a USB drive or something. This also isn't a counterpoint to your argument that Windows tech is better supported given its origins, but it might be useful for some people depending on their intended use.
That's a good tip, and I do use exFat on some pendrives. But due to the lack of journaling, and its buggy performance on macOS ( https://www.linkedin.com/pulse/exfat-file-system-save-henk-s... ) I wouldn't recommend it for long-term use on any fixed drives with data you care about. My research lead me to conclude that NTFS implementations are the least buggiest non-native file systems on Linux and macOS.
Interesting, I wasn't aware of the exFAT issues on MacOS. I can't remember exactly the last time I tried to use a USB drive like this, so it's possible that it might be further back than I thought and it was when NTFS didn't have write support out of the box for me on Linux.
Yes, interestingly I remember buying portable hard drives 20 years ago that were formatted as FAT or some variant (I don't remember which one exactly).
Last time I bought a portable hard drive it was formatted as NTFS.
If MS abandons WinNT, then people will likely continue to use the existing versions of Windows which are out there for any existing software (just as people continue to use MS-DOS and Win 9x for old games and software).
As for new software - I think it's open to debate just how much new Win32 software will be created after a hypothetical abandonment by Microsoft of Windows.
> That's much more relevant to industrial applications where some device is controlled by an ancient computer because the vendor originally provided drivers for NT 5.0 or 5.1 which don't work on anything modern.
In most of those applications, you just leave the computer be and don't touch it. In some cases (especially medical devices) you may not even be allowed to touch it for legal/compliance reasons. If the hardware dies, you most likely find the exact same machine (or something equivalent) and run the same OS - there are many scenarios where replacing the computer with something modern is not viable (lack of the correct I/O interfaces, computer is too fast, etc.)
If there were software bugs which could impact operations, they probably would have arisen during the first few years when there was a support contract. As for security issues - you lock down access and disconnect from any network with public internet access.
All that assumes that ReactOS is a perfect drop-in replacement for whatever version of Windows you are replacing, and that is probably not a good assumption.
In my experience, things like ReactOS would have been more useful in parts of the world with let's say a less thorough approach to things like compliance.
A factory has a CNC machine delivered fifteen years ago that's been run by the same computer all along. The computer eventually gives up the ghost, the original IT guy who got the vendor's drivers and installed them on that computer with an FCKGW copy of WinXP is long gone. Asking the current IT guy, the easiest solution (in a hypothetical timeline where a usable ReactOS exists) is to take the cheapest computer available, install ReactOS, throw in drivers from the original vendor CD at the bottom of some shelf and call it a day.
We might have to agree to disagree here, but I think the scenario where the IT guy uses XP and "finds" a license for it is the approach I would take if I was put in this situation. If the vendor for the CNC machine certified/tested their machine against Windows XP, and does not offer any support for new operating systems, I would be very reluctant to use anything else - whether it is another version of Windows which could accept the same drivers, or an open source clone. Again, I'm assuming that ReactOS manages to be a perfect clone, which is may or may not be in practice.
As someone who generally uses metric units, but grew up around English Imperial units - if an American says that a person weighs a certain number of pounds, I need to convert to stone and pounds in my head in order to get a meaningful mental model of how much that person weighs.
> a) never plan on learning and just care about outputs, or
> b) are an abstraction maximilist.
As a Claude Code user for about 6 months, I don't identify with either of these categories. Personally I switched to Claude Code because I don't particularly enjoy VScode (or forks thereof). I got used to a two window workflow - Claude Code for AI-driven development, and Goland for making manual edits to the codebase. As of a few months ago, Claude Code can show diffs in Goland, making my workflow even smoother.
My only gripe with the Goland integration is if you have multiple terminals open with CC, it will randomly switch to the first terminal for no apparent reason. Then, if you aren't paying close attention, you prompt the wrong instance.
After reading this article, it is unclear to me what "The Myth of the Thinkpad" is. The author laid out a series of desirable attributes about Thinkpads, and proceeds to explain how they are all true. We are reminded that all these desirable qualities arise from business realities, which is generally true for any desirable attribute in a commercially successful product. In the end, there's a conclusion that Thinkpads are a good choice, but that they are not "magic", as if this was something that needed explanation.
To put it bluntly, the author thinks that people who praise Thinkpads are dumb. They think we all believe that Lenovo/IBM listened to us lowly nerds and forged a one of a kind Thinkpad to satisfy our dream of a serviceable and durable machine, and not purely to make money.
> May be better with a native PCI-e or PCI HBA and 700W power supply and a junker ATX Linux machine to provide network shares.
Agreed. Even if it's possible to get a combination of adapters to allow a SCSI interface to be attached to a Mac (and assuming the correct driver support is present), I think getting an old PC and an old SCSI adapter card may be cheaper.
reply