Talk:GNU variants

Merge

From Wikipedia:Merge: Wikipedia is not a dictionary; there doesn't need to be a separate entry for every concept in the universe ... If an article is very short and cannot or should not be expanded terribly much, it often makes sense to merge it with an article on a broader topic.

I don't think this article is going to become large any time soon. It will probably never grow large, but if that does happen, I'm certain enough that it will not be in the near future. There is a page about a similar topic: GNU/FreeBSD, and that page isn't much bigger than this one. I would like to suggest merging these two articles, but to what name? The unified article should remain open ended so that if a GNU system with the OpenBSD kernel comes into existance, it will also fit on the unified page. "GNU Operating Systems with BSD kernels"? "GNU Operating System variants" or "GNU Operating Systems with alternative kernels" and then in the opening line say "for linux-based variants, see GNU/Linux? Suggestions very welcome. Gronky 15:01, 11 November 2005 (UTC)

Normal GNU (with the Hurd) and GNU/Linux are very common and deserver their own page. Let's just make this an article about any minor GNU/$kernel. These would be GNU/kFreeBSD and GNU/kNetBSD and maybe GNU/OpenSolaris aswell, though that one might be large enough to justify its own article. (Nexenta is quite active, from what I understand, but I'm not sure about that.) Geronimooo 15:40, 27 November 2005 (UTC)

So I made this merge -- mms 13:42, 8 January 2007 (UTC)

What do you have to support your assertion that GNU with the hurd is common? debian popcon reports just 11 systems using hurd compared to nearly 100 using kfreebsd and over 100K using linux. Further kfreebsd has actually made it into a debian stable release (and is on-target for continued inclusion) while hurd is looking like it's going to get kicked out of the official archive (having never made it into a stable release) unless the porters SERIOUSLY pick up their game in the near future. Granted debian isn't the only hurd distro but even so things look far more rosy for kfrebsd than for hurd to me. Plugwash (talk) 15:56, 6 December 2011 (UTC)

Sorry didn't spot how old these posts were, Kfreebsd was less common back then. Still I haven't seen any evidence that hurd was EVER common. Plugwash (talk) 15:57, 6 December 2011 (UTC)

Debian GNU/NetBSD

Needs a remark on how usable it is to be added --anon.

Nexenta OS

Binary drivers

»This makes it far simpler for hardware manufacturers to develop device drivers for it while protecting their IP rights.«

This statment is a bit unbalanced. It presents binary drivers as an unalloyed benefit, and offers a surious justification: hardware manufacturers don't want binary drivers because there're easier to write, it's purely to avoid releasing code. Could there be a link to, say binary-only driver, to present some of the disadvantages of binary drivers. 203.214.125.30 09:50, 31 December 2006 (UTC)

I changed it to »This makes it legal for hardware manufacturers to release device drivers without the source.« -- mms 13:42, 8 January 2007 (UTC)

revert

Hey, Xaero Vincent, as I can see you solely contributed to Nexenta OS till now. You didn't recover the article there yet and I don't want an edit war here but I think your revert and the new text about GNU/OpenSolaris is wrong. AFAIK Nexenta OS is a new name for GNU/Solaris which is sometimes also named GNU/OpenSolaris. On http://www.gnusolaris.org you have all 3 names. Now we have two questions:

1. Is Nexenta OS GNU/Solaris and GNU/OpenSolaris and therefore a GNU variant?
2. Is the section about Nexenta OS worth to be an article?

The first question I would answer clearly yes. For the second question I don't have a strong opinion but I would say not yet. -- mms 17:57, 11 January 2007 (UTC)

I think Nexenta OS deserves its own page; in fact, until recently it did have its own page. I think there's enough info to warrant its own article.

I actually don't know if this article is really warranted. These operating systems are only really related in a very loose sense, that being that they are all built around GNU software in some way, but that's a loose association, and the grouping here doesn't make a whole lot of sense to me. – Mipadi 18:23, 11 January 2007 (UTC)

As you can read in the archives there has been discussions to merge all those *GNU/*BSD articles. But under which name? I think »GNU variants« fits it well. At this time the article is not much more than the product of a merge. Read GNU to get an idea why GNU is such an unique operating system. Without it there would be no "Linux" and there would be no free BSDs -- except there would have come another person like RMS around. And I think it's no accident that all other variants than GNU/Linux described in this article are a Debian project or Debian-based projects. Debian generally values freedom still much more than all "Linux distributions" which one can see in naming theirs GNU/Linux. And even so I oppose calling GNU/Linux Linux I accept that the developers of GNU/Solaris or GNU/OpenSolaris have chosen to call their operating system Nexenta OS. At least it isn't named after its kernel. As long as they don't claim it is "Linux-based" (but GNU-based) it's okay for me. It is legal to do so. And the history is totally different. There were GNU users before Linux came out. Then Linus Torvalds developed the first versions of Linux and changed it license so it could be combined with GNU. The early GNU/Linux users knew they are running a GNU variant. That is why they demand the name GNU/Linux for the thing which currently is in the article Linux. Also GNU/Linux is by far more important than Nexenta OS and so I'm more relaxed here. See also GNU Users Who Have Never Heard of GNU -- mms 19:31, 11 January 2007 (UTC)

I'm not exactly sure what or who you are responding to. You indented underneath my comment, but you don't seem to really be addressing my statement. You gave an interesting, if unnecessary, history of GNU (and, I might add, seemed to assume I know nothing of the GNU project or the development of the Linux kernel, and how they relate), but I'm not sure what you are addressing, or who it is directed at. – Mipadi 20:17, 11 January 2007 (UTC)

I was responding to your comment but drifted away a bit. I also addressed your last edit to the article. -- mms 21:10, 11 January 2007 (UTC)

My last edit was this, which is just a bit of cleanup, aside from changing the wording of the line about GNU/Linux a bit. But it's not cut-and-dried that it's wrong to refer to such a system as Linux—that's why there's a controversy, after all. – Mipadi 22:27, 11 January 2007 (UTC)

Ging?

Ging redirects here. This is useless until someone can add what "Ging" actually means to the article. (And no, "ging is not ging" is not enough on its own!) 86.136.255.69 (talk) 21:08, 15 February 2008 (UTC)

Can you add something to this article to say what Ging is? If not, I hope someone else will - but if you know something, please help. --Gronky (talk) 09:16, 16 February 2008 (UTC)

A Debian/kFreeBSD LiveCD, apparently: http://glibc-bsd.alioth.debian.org/ging/ EdC (talk) 00:09, 21 February 2008 (UTC)

Next time you need info on Linux etc., try [1] --Thinboy00's sockpuppet alternate account 19:30, 8 May 2008 (UTC)

Debian GNU/kFreebsd

I believe this is currently bootstrapping itself in the main archive. This would mean it can release with Squeeze. http://lists.debian.org/debian-devel-announce/2009/04/msg00001.html Not an official release yet, but almost there. kgoetz (talk) 07:31, 3 June 2009 (UTC)

The link to the "Main Article" for Debian GNU/kFreeBSD is looping back to this article. Where should it lead to? Debian's article, maybe? -- Zeratul2k. —Preceding unsigned comment added by 190.182.80.178 (talk) 19:10, 16 October 2009 (UTC)

Gentoo/Alt

• Gentoo/Alt is a project to provide an operating system based on a GNU userland managed by Portage to with different kernels including FreeBSD, OpenBSD and NetBSD.

This section is incorrect. Gentoo/FreeBSD (not sure about the rest of Gentoo/Alt) uses the FreeBSD user-land. See The official page. It does use the GNU tool-chain, though. --Theriddle (talk) 00:36, 28 June 2009 (UTC)

I agree with Theriddle about the Gentoo/Alt GNU Issue. It's not GNU. It's still primarily a BSD Userland. Just because it contains a GNU toolchain shouldn't make it automatically a GNU variant. I would request to remove it from this article or atleast mention that it only has the GNU toolchain. --Fader05 (talk) 12:00, 9 August 2009 (UTC)

Redundant "main article" link

The section "FreeBSD variants" is followed by a link to the full article "Debian GNU/NetBSD." However this link leads right back to this article so I have deleted it. 86.41.80.64 (talk) 10:35, 29 October 2009 (UTC)

Nexenta_OS says, that

Nexenta OS is the first distribution that combines the GNU C library and userland with the OpenSolaris kernel

while GNU_variants article says, that

Nexenta OS is the first distribution that combines the GNU userland (with the exception of libc

One of them is wrong. --Norayr (talk) 10:49, 30 March 2010 (UTC)

What is active development?

Hi, Just wondering if there is a standard or active development on wikipedia? What does the project have to show to be active?

kgoetz (talk) 22:30, 11 July 2011 (UTC)

As expected, there can't be a clear-cut definition of "active development", and I don't think there is any policy/guideline on wikipedia, however, here is my take: if there are a lot of people working on it a a relative short timeframe (1 week/month), making lots of improvements, than that could be a "metric". An even more specific metric could look into the number of commits/changes to source-code repositories available on the internet. If a project makes very frequent releases, it might also be an indication of active development (but not necessarily). Inversely, if a project has not shown any release or any type of development in, for example, the last year, it might be hard to "prove" "active development". In concrete cases, I think projects like Firefox, Fedora or Ubuntu clearly meet this definition. Regarding Debian GNU/Hurd and my edit, since the last version was from 2009 and I saw no evidence of active development (in fact, no evidence of any recent development at all), I felt it was incorrect to call it under active development with such a lack of information. However, after reading your comment, I found out the news page which at least seems to indicate the project is not dead. Anyway, if you are familiar with the project and think it is actively developed, feel free to revert my edit, that will not be a problem, but if you do that, I ask you to use wording like "As of June, 2011, this project in under active development", so this way information remains accurate in the future. --SF007 (talk) 23:22, 11 July 2011 (UTC)

Sorry about the delayed reply, I forgot i'd posted here. Hurd hasn't been released with Debian yet, but may be released as a technology preview with Wheezy. I feel it is under active development, as its development list receives regular posts with patches for/discussion of portability. I guess as its the low end of the 'active' spectrum (eg, active but not very active) i won't be too fussed with what happens wording wise. -- kgoetz (talk) 09:27, 8 August 2011 (UTC)

Debian doesn't split out contributions by target architecture/OS so it's difficult to tell how much activity there is unless it happens to show up on a mailing list. Still the stats on buildd.debian.org paint a picture of sporadic imrovements though they also show that they are a long way from meeting the release critera Plugwash (talk) 16:07, 6 December 2011 (UTC)

Does complete GNU systems belong in an article about GNU variants?

Hasn't anoyone here yet reflected about Arch Hurd and Debian GNU/Hurd not being GNU variants, but just actual ordinary GNU systems (besides me in September, but my edit was reverted as vandalism)? Or is the definition removed in this edit wrong? What does the term "GNU variants" actually mean, and is it actually used at all? Or is this just a convenient way of collecting different systems into one topic (in that case, what should the boundaries of that topic be)?

So, to be really clear: if a GNU variant is a GNU system not using the Hurd kernel, then Arch Hurd and Debian GNU/Hurd does not belong here, since they do use the Hurd (and therefore are just ordinary GNU systems). But what do we actually mean with GNU variants? Do we have any sources supporting that the term is used in the broader or narrower sense, or even in any sense at all? I can't see that the current footnotes supports this statement. Kindly, 78.82.251.51 (talk) 10:49, 13 August 2011 (UTC) (user:flinga on the Swedish language Wikipedia)

The statement that it's a term used by the FSF was added here. This was given sources in this edit, but those sources does not support that it is an established term, since the sources does not seem to contain the term. If it isn't established as a term, we should not state that it is. And if we want an article about Non-Hurd GNU systems, maybe we could call it something else (for example, "Non-Hurd GNU systems")? Or do we want an article about all possible GNU variants, both Hurd and Non-Hurd variants? Kindly, 78.82.251.51 (talk) 11:35, 13 August 2011 (UTC)

Afaict despite being a project to create a unix-like operating system the GNU project has never actually officially released one (though it looks like there may have released a "demo image" for hurd which our GNU article appears to be dubiously clamining as a "release of GNU"). They have released many of the components for an operating system but they have left it up to others to perform the actual task of combine those components with other free software to create the operating system. Plugwash (talk) 01:53, 6 March 2012 (UTC)

And, hence, all releases of GNU are GNU variants, is that the argument? It's like it would be parallell to the expression "Linux distribution" (although "system-centered", not "kernel-centered")? Well, we would still need sources for claiming that this is a term.

My argument is that today we state that GNU variants is a term used by FSF, mening any system using the core userland from GNU, but we do not have any source that supports this statement.

We could avoid saying that it is a term used by the FSF (since it hardly seems to be), and instead just have an article about variants of the GNU system (regardless of kernel as today). We could as well, if we wish, call the article "GNU systems", or any other fitting term. We could also narrow the scope and just include alternate versions of GNU, not running the GNU-kernel (Hurd) (which I thought was the original intent with this article). Or we could merge it (perhaps not neccessary) in the GNU article. That would all be correct. But we should not say that this is a term used by FSF if that's untrue, and we should be clear about what's in the scope of the article. Kindly, 78.82.197.237 (talk) 22:25, 9 August 2012 (UTC) (user:flinga@svwp)

The two problems are: 1) is this an article about "variants of GNU" (where variation to me sounds like some sort of alternation; with an alternative kernel), or about "versions of GNU" (regardless of the kernel)? 2) there are no source supporting the claim that this is a term at all. 78.82.197.237 (talk) 22:35, 9 August 2012 (UTC)

This article is about variants. Versions of GNU are documented in GNU. The "Hurd variants" section should be removed or merged into the GNU article. Gronky (talk) 12:01, 2 May 2013 (UTC)

debian hurd

I'm not putting this in the article right now because it would have to be worded very carefully to avoid stepping on OR and/or SYNTH but the bottom line is that at this point it would take a miracle for hurd to released with debian wheezy. Their build percentages are still far from acceptable, they haven't been in dialog with the release team, they haven't been included in testing even as a "broken and fucked arch"*.

* New architectures generally get added to testing as a "broken and fucked arch" (debian terminology not mine but it basically means they don't impact testing transitions for other architectures) and then later get switched to being a release architecture when their state in testing is considered good enough.

-- Plugwash (talk) 02:08, 6 March 2012 (UTC)

I do share your expectation, especially because the Hurd developers are essentially 5-10 part timers. To even complete java implementation, their DDE driver project and maybe get only one simple GUI running without major bugs until the end of 2012 seems excessively ambitious to me (I wrote the factual part of this in the article, btw). However, the ambition is there. The developers publish snapshot dvd images every quarter year, they keep porting packages and they still work on the hurd. So even though probably almost everyone expects this to be just another instance of "almost critical mass" which the hurd seems to see every 5-6 years in its life cycle, this is not something that can be written into the article unless someone FOSS- or Debian-notable says it out loud. -- 178.25.83.219 (talk) 09:47, 6 March 2012 (UTC)

Hurd variants are not "variants"

The original system is GNU, which uses Hurd. GNU/Linux is a variant because it's not exactly GNU.

GNU systems using Hurd are already documented in the GNU article. Can I merge the "Hurd variants" sections from here to some part of the GNU article? Gronky (talk) 11:55, 2 May 2013 (UTC)

Would a system comprising the Hurd kernel and a non-GNU userland not be a GNU variant?

The current definition of GNU variant at the beginning of the article doesn't allow for a hypothetical combination of GNU Hurd and a non-GNU core userland to be considered a GNU variant. --isacdaavid 02:05, 28 April 2014 (UTC)

Yes, I believe that combination would not be GNU, and easily found sources would back that up. The kernel (Hurd) was the "last" part of GNU. The compiler and C library one of the first. The C library would be a system library that would be "core userland", that is GNU. The compiler "application" software. comp.arch (talk) 13:57, 22 May 2014 (UTC)

The time has come to discuss this, not merely as a hypothetical, I added in a paragraph for Alpine/Hurd which takes the non-GNU Linux using Alpine, an Unix-like operating system, but now adds in the Hurd kernel in place of Linux. Basically this is to me such an edge case where it may be difficult, the system does feature a distribution of the GNU software and the kernel is noteworthy for an operating system, but is it important enough, basically in this case the Hurd kernel is the one that gets added in at a later time much like how the Linux kernel gets added in at a later time to GNU/Linux variant GNU distros. Basically, I argue, and this is my line, that the historical lineage is what determines the OS family, Alpine does not really as a system which uses no GNU software derive from the GNU family, (unless it historically has used GNU software which was later removed), but now Hurd is added in.

Either A. Alpine did once derive from the GNU system, removed GNU software, now Hurd is added in, in which case either that's important and it's a GNU descendant, or it isn't important and it isn't one. Or B. Alpine is not a GNU derived OS, but Hurd now gets added in as a Unix-like kernel for an OS, which is a component, Alpine is thus still just a Unix-like system.

In any case I think there should be a mention of Alpine/Hurd regardless, somewhere, even if to point out it fails the criteria for being a GNU distro, I believe the connection to the kernel is noteworthy so it should be mentioned to that capacity even if it is determined to be a non-GNU system. Basically if someone can make good arguments, then the paragraph I wrote can be moved elsewhere in the article, maybe even a new section? Anyway I hold no strong opinion on this at this time, and I post this now hoping to merely start the discussion. Softwareperson1000 (talk) 15:12, 2 May 2026 (UTC)

GNU-Darwin

Should GNU-Darwin be listed in this article? I've never been too clear as to what exactly its status is, and what it "counts" as... Cooljeanius (talk) (contribs) 01:23, 9 December 2014 (UTC)

Support for GNU-Darwin as a GNU variant is based on the fact that it certainly is a distribution of the GNU software in an operating system environment, CDs were distributed featuring the GNU environment, and that is in the taxonomic sense a "distribution of GNU", where I don't differentiate for the word 'distribution' between operating systems versus non-operating systems kinds, distributions of operating systems are just operating system distributions to specify that kind of distribution. The question is about does "a distribution of GNU" count as an operating system distribution, it is some GNU distribution, but what kind?

But then they did also sell hard drives containing a full complete Darwin installation + the relevant GNU software to make it a complete GNU-Darwin system distribution, so there really was a complete GNU-Darwin distro.

For sure GNU-Darwin is more a GNU distro than the Cygwin Windows NT one is at least. So if one of the two didn't belong here, start with the Windows NT. Perhaps a separate category on this article can be made for complete OS distributions, versus the software distributions intended for use in another system environment? Softwareperson1000 (talk) 15:23, 2 May 2026 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified 2 external links on GNU variants. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20070711071736/http://www.gnusolaris.org/gswiki to http://www.gnusolaris.org/gswiki
• Added archive https://web.archive.org/web/20110725012919/http://www.archhurd.org/ to http://www.archhurd.org/

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 22:18, 6 January 2017 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on GNU variants. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20081120014944/http://www.superunprivileged.org/hurd/live-cd/ to http://www.superunprivileged.org/hurd/live-cd/
• Added {{dead link}} tag to irc://irc.oftc.org/debian-kbsd

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 16:34, 9 October 2017 (UTC)

New Section "Impact" idea

I had an idea for a new section for "Impact", maybe have something about Sun Microsystems having hired the Debian founder Ian Murdock wwho would work there on project Indiana that would bring about the OpenSolaris distribution and later illumos. I'm also thinking of reviving the old Windows Subsystem for Linux I removed in an earlier edit, now back into that new category. Maybe some general inspiration the GNU Project has had on the BSD systems which is also something I've heard said, could be useful.

Also another thing, now that I've added the category for GNU software used in other operating systems I was wondering of how broad should the scope of that be, as in, should Haiku warrant a mention since Haiku makes use of a notable amount of GNU software, it could be argued. Not enough GNU software such that I would argue it is a GNU operating system, and it's historical lineage doesn't allow for that, but maybe a mention of it could be made still? If so how about other operating systems? Would it be useful to add various operating systems the GNU project has had an effect on in their inclusion of some substantial amount of GNU software, I know many illumos derived projects at least use lots of GNU software so that should be included. Some threshold for a meaningful amount of GNU software would need to be had. And the mentions of the various systems could be very broad and generic, and vague in the sense that they don't have to go on to say much about those operating systems in particular, just that some x GNU software is used there and there, maybe? I'm not going to go ahead with this yet, unless someone says something about it being a good idea, or until I can form a more coherent plan. Softwareperson1000 (talk) 12:40, 3 May 2026 (UTC)

More ideas for new sections:

1. Should there be a History section on this article also? It would probably duplicate too much with the article GNU that also has one, but it's something to consider.

2. It's probably a good idea to have a section explaining the basic concepts of Unix-like design that the GNU family of operating systems follow. As all the GNU variants I know of are all entirely Unix-like. Even if there was some that weren't Unix-like they would be very fringe and obscure indeed and still because most are, and the GNU Project was itself started as one, having the Unix-like design mentioned on this article as it's own section is very much appropriate, if it is to be added.

3. It could be useful to have a brief section on the free software movement then linking to it's article as the 'Main article' because a lot of these GNU variants are predominantly free software projects. And this article has historically focused on distros that are explicitly FSF and GNU Project endorsed even before I showed up here, so it seems relevant to make a mention of the free software movement as it's own section. I already added it into the 'See also' section earlier, but that may not be enough.

4. Further on the topic of the free software movement, given this article is the GNU variants, it is also probably useful to have a section on the GNU manifesto. The section could be basically duplicated from the article the GNU Project more or less identical I think, because both articles this one and that, are tied to the manifesto in a similar way. Is that a good idea? Is there any rule or policy on Wikipedia against duplicating content from one article to another like this that I don't know of? Anyway I don't know if I'll get to it anytime soon, someone else can feel free to do that. Softwareperson1000 (talk) 14:37, 4 May 2026 (UTC)

I had an idea for a new section for "Impact", maybe have something about Sun Microsystems having hired the Debian founder Ian Murdock wwho would work there on project Indiana that would bring about the OpenSolaris distribution and later illumos. To what extent does Project Indiana/OpenSolaris/Illumos owe anything other than the idea of being free software to that? "GNU userland atop the SunOS 5.x kernel" is an interesting project, but that project is already covered in GNU variants § OpenSolaris and Illumos kernel.

I'm also thinking of reviving the old Windows Subsystem for Linux I removed in an earlier edit It's back there, but to what extent was WSL1 just "GNU/NT kernel" with the NT kernel having a Linux system call handler added to it (as some \*BSDs have had), which might be more appropriate for GNU variants § Windows NT kernel?

Would it be useful to add various operating systems the GNU project has had an effect on in their inclusion of some substantial amount of GNU software Unless they can reasonably be characterized as "GNU variants", no. There's plenty of room for that in, say, GNU or GNU Project. Guy Harris (talk) 22:01, 13 May 2026 (UTC)

It's probably a good idea to have a section explaining the basic concepts of Unix-like design that the GNU family of operating systems follow. I don't see why; does it belong here and in GNU and in GNU Project and in every BSD page and in the macOS page and in pages for anything that produced a Unix-like operating system without using AT&T code?

It could be useful to have a brief section on the free software movement then linking to it's article as the 'Main article' because a lot of these GNU variants are predominantly free software projects. Ditto - I'd say discussion of that movement belongs in GNU Project, but stuff downstream of that project, such as GNU and this page, don't need to discuss it.

Further on the topic of the free software movement, given this article is the GNU variants, it is also probably useful to have a section on the GNU manifesto. The section could be basically duplicated from the article the GNU Project more or less identical Same. No need to repeat all that stuff in multiple pages with the name "GNU" in them. Guy Harris (talk) 22:10, 13 May 2026 (UTC)

What I meant by the Sun and OpenSolaris thing is that, if my understanding is correct, the reason for what brought about the Project Indiana was the widespread success and prevalence of the GNU based operating systems like Debian, which compelled Sun Microsystems to compete against that and release OpenSolaris and it was lead by none other than the Debian founder himself. But my recollection of such history may be flawed so I may be wrong. It's something I recall hearing at some point someone say. So the basis for it's relevance here, is because it is the Impact that the GNU had in the world, so it's relevant to the article (or I suppose to the various other articles as you later point out also exist on the GNU Project).

I am confused as to what you are alluding to about the Linux system call handler, what has the Linux kernel got to do with the GNU variants? BTW I was the one who put WSL back into there to the new Impact section I added as per what I said here.

Since the post you now replied to here, I have decided it is necessary to have a mention of the Unix-like design, because that is absolutely essential to provide crucial context for why, and how, these systems even work and exist to begin with. What are the GNU system distribution variants, if not Unix-like systems where the modular design and the separation of the kernel allow for the kernel to be swapped out, thus the GNU variants? I say that is essential context and information for the article, as that is literally the main focus of the article. Otherwise the article assumes the reader is familiar with the Unix-like design and the fact that the kernel can be swapped out as the article implies (if not explicitly explain it). You can't assume the visitor / reader knows what a kernel is, so explaining the nature of the Unix-like systems and the separation of the kernel explains why there are many GNU system distributions with different kernels, GNU variants

As to the inclusion of duplicate mentions of things already mentioned on multiple GNU focused articles, what you said makes sense, however, as it stands I say there are too many GNU articles, See here for my case here that either the article GNU be removed or repurposed because it is an aimless mess of an article that doesn't really serve a purpose among 3 GNU articles + all the other specific GNU software related ones like GNU packages or whatever etc. Softwareperson1000 (talk) 12:45, 14 May 2026 (UTC)

I am confused as to what you are alluding to about the Linux system call handler, what has the Linux kernel got to do with the GNU variants? What I said was "with the NT kernel having a Linux system call handler added to it", which isn't about the Linux kernel, it's about the NT kernel. See WSL System Calls. The point was that version 1 of the Windows Subsystem for Linux was somewhat like "GNU/NT", in that it ran a GNU/Linux userland atop the NT kernel. It wasn't "GNU/NT" in the sense of an OS whose primary interface is that of GNU/Linux, it's more like the Linux binary compatibility offered by some *BSDs with their Linux system call emulators.

You can't assume the visitor / reader knows what a kernel is, so explaining the nature of the Unix-like systems and the separation of the kernel explains why there are many GNU system distributions with different kernels, GNU variants The original GNU was intended to be a Unix-like system, which is where "Unix-like" enters into it. The ability to run GNU userland code, with some modifications, atop kernels other than what ended up being, I guess, the GNU kernel - GNU Mach including various servers - is mainly due to GNU libc having been written to be portable to other OSes, and thus due to being able to support the system call interfaces offered by various OSes. The GNU C library includes routines such as open() that are implemented, for most OSes supported by the library, as system call stubs. The stubs differ from kernel to kernel. The rest of the GNU userland is implemented atop the UNIX API provided by the GNU C library. That particular type of modularity is not unique to Unix-like systems; the Win32 API is implemented atop two different core OSes, the Windows 9x family and the NT family, with very different low-level implementation details, to allow (some) binaries to run atop both Windows 9x and Windows NT (they either have to avoid the Unicode APIs or use the library Microsoft provided to support those on Windows 9x) by calling routines from, say, kernel32.dll, with the Windows 9x version of kernel32.dll doing whatever is necessary to implement, for example, CreateFile() atop the Windows 9x core and the Windows NT version of kernel32.dll doing whatever is necessary to implement those routines atop the Windows NT core (not just the kernel - CreateFile() passes, as I understand it, through servers such as csrss.exe, although ReadFile() and WriteFile() are, as far as I know, wrappers over the NtReadFile() and NtWriteFile() system calls). Guy Harris (talk) 23:31, 17 May 2026 (UTC)

So the Windows NT thing confusion is probably because the way I've understood it, is that there was no actual distribution of GNU software involved in the WSL software Microsoft provides, which means as just a compatibility layer with the Linux kernel, it does allow running GNU software built targeting Linux but now on Windows, however the actual WSL software itself that Microsoft provides, does not contain any of GNU code (to my knowledge anyway) so WSL itself can't be a variant of GNU. Which would mean then that it's those third party distributions of software as the disk images which then contain the GNU code, which would then have to be the GNU variants (if to be included in the article). I think such details aren't explained well (at all) in the article, the relationship between those disk image things that the WSL compatibility layer runs versus the WSL software itself.

I still think the Unix-like information is useful to explain for the reasons I stated, unique to Unix-like design or not. It serves the article to provide context and a basic explanation of how those many GNU variants relate to each other. As it stands, the page just lists the variants but doesn't explore what are really (they are just GNU variants, whatever that is) and importantly what do they have in common. How they differ from each other in relation to what they have in common, is the same as in how the kernel relates to the rest of the system. Softwareperson1000 (talk) 17:25, 18 May 2026 (UTC)

So the Windows NT thing confusion is probably because the way I've understood it, is that there was no actual distribution of GNU software involved in the WSL software Microsoft provides, which means as just a compatibility layer with the Linux kernel Yes, similar to the Linux compatibility layers that some BSD systems provided. Atop that you'd put Linux libraries and executables; I think Microsoft may have provided those binaries for some distributions, but they didn't develop them.

WSL2 runs a complete Linux distribution, kernel and all, inside a lightweight virtual machine, as per this WSL2 announcement.

So WSL1 is more like "running Linux libraries and executables on top of pickyourBSD's Linux compatibility layer" than like GNU/kpickyourBSD.

It serves the article to provide context and a basic explanation of how those many GNU variants relate to each other. The way that the notion of a Unix-like system relates to GNU variants is that they're variants of a Unix-like system. I don't see any other way in which it relates. Guy Harris (talk) 21:55, 18 May 2026 (UTC)

I'll leave the issue of the WSL to other people to decide what to do with, because my familiarity with the Windows systems is too limited to make further comments, I've now explained how I saw it.

I will also respond to what you said earlier, I would actually prefer if all operating systems derived from Unix-like design had a mention of that Unix-like design in the article. I think that's useful and relevant information in general. But I don't really have anything more to say on this issue of how GNU variants relate to the Unix-like design, I've made the arguments which I still stand by. Whether reiterating them a few more times could change whether you see it or not is probably not necessary. Softwareperson1000 (talk) 10:54, 19 May 2026 (UTC)

1) What is a "Unix-like design"? Does it mean "an API and command-line design that is more or less similar to the various Unix systems from Bell Labs"? If so, I think that belongs in Unix-like - or, in detail, in Unix, with Unix-like referring to it - and that, given that the Wikipedia is a wiki, and that a wiki uses hyperlinks, the way to introduce that information is by hyperlinks to Unix or Unix-like on pages about Unix-like systems, rather than repeating it on those pages.

2) The GNU systems don't relate to it beyond "we wanted to build a free-software operating system that's compatible with Unix, and better in some technical ways". In the GNU Manifesto, Stallman says, in the "Why GNU Will Be Compatible with Unix" section, "Unix is not my ideal system, but it is not too bad. The essential features of Unix seem to be good ones, and I think I can fill in what Unix lacks without spoiling them. And a system compatible with Unix would be convenient for many other people to adopt." I wouldn't call that "damnation with faint praise", but it speaks not at all about the "Unix philosophy", for example.

The "What's GNU? Gnu's Not Unix!" section says

GNU will be able to run Unix programs, but will not be identical to Unix. We will make all improvements that are convenient, based on our experience with other operating systems. In particular, we plan to have longer file names, file version numbers, a crashproof file system, file name completion perhaps, terminal-independent display support, and perhaps eventually a Lisp-based window system through which several Lisp programs and ordinary Unix programs can share a screen. Both C and Lisp will be available as system programming languages. We will try to support UUCP, MIT Chaosnet, and Internet protocols for communication.

which speaks about adding ideas from non-Unix systems - the references to Lisp probably come from Stallman's experience at the MIT AI Lab and with Lisp machines, and I'm not sure where the "file version numbers" idea comes from (I don't think Stallman worked much with RSX-11 or VMS, but the Incompatible Time-Sharing System (ITS) may have had them. I suspect many adherents to the "Unix philosophy" would not support version numbering for files.

So, if I'm not seeing reasons for it being important to discussion of GNU systems, perhaps that's because those reasons aren't there in anything the GNU people themselves have said. Guy Harris (talk) 19:21, 1 June 2026 (UTC)

And if the design your thinking of is "a system where all components are replaceable", where you could, for example, replace the kernel, that's 1) not an inherent property of Unix-like systems and 2) not unique to Unix-like systems, so it's not a "Unix-like design".

Note also that replacing the kernel may mean replacing parts of userland. If the system call interface is different, the system call wrappers in the system library might have to change, for example, so it's not as if replacing the Hurd with the Linux kernel was just "slide out one kernel and slide in the other". The GNU C library was built to support multiple OSes, so it was already prepared to support different kernels with different system call layers, so that didn't require much work, but it did involve adding Linux-kernel support. Guy Harris (talk) 20:10, 1 June 2026 (UTC)

But here is your mistake, you assume we are talking about the GNU Project broadly and in general, however this is not an article about the GNU Project, this article is the GNU variants, which is a specific article that covers GNU systems where the kernel is swapped out for another. This it is in direct relation to the contents of the article to explain the differences between the systems, as in, these are the systems of GNU variants and their difference is the swapping out of one kernel for another, Unix-like operating systems feature such a distinction between the kernel and the rest of the operating system to enable this practice of swapping out the kernel, hence GNU variants. Because it is directly describing the contents of the article, Stallman's goals in targeting Unix as the model to build GNU around as historical facts aren't really all that relevant to my argument right here right now, where the fact is, GNU was made as Unix-like regardless of the reasons and what led to them, here it is, we have an Unix-like design and here are the GNU variants Softwareperson1000 (talk) 10:07, 2 June 2026 (UTC)

> "a system where all components are replaceable", where you could, for example, replace the kernel ... not an inherent property of Unix-like systems ... not unique to Unix-like systems, so it's not a "Unix-like design".

The key detail is the interchangeable nature of the kernel specifically and in relation to the GNU variants. The kernel being interchangeable is a manifest property of it's separation from the rest of the system, whether it was actually and intentionally designed in that way specificially to be a replaceable component is not the point, it's modular design as an independent module in the system as per it's separation from the rest of the system is what causes it to be replaceable inherently, and the GNU variants prove this can be done. Which means whether intentional or not, the effective result is such that the kernel is an interchangeable component, because, it is separated from the rest of the system. The rest of the Unix-like modularity does tie into that, does some Unix specification demand the systems be modular, who cares, in practice that's how the systems work and that's what the GNU variants as an article are dealing with.

See the documentary The UNIX System: Making Computers More Productive https://archive.org/details/theunixsystem

The kernel is clearly established to be a separate component within the operating system, and most of all, not the operating system, the "operating system" is a whole that contains the kernel, it does not consist of the kernel with the optional addition of other things. Softwareperson1000 (talk) 11:40, 2 June 2026 (UTC)

But what is an Unix-like design, as in is there some definition or a specific criteria for it, I don't know, I can point to you examples of several operating systems which are widely recognized as being Unix-like and I can point to some concepts of the file systems and Unix philosophy and such things, but I don't recall anyone writing a widely accepted specific definition for what counts as Unix-like. But here is the fact, there clearly are lots of operating systems out there in the world, that are seen as Unix-like. And Unix-like operating systems feature the separation of the kernel and the rest of the system, and GNU variants are systems where one kernel is swapped out for another whilst the predominant userland features remain Softwareperson1000 (talk) 10:13, 2 June 2026 (UTC)

Unix-like perhaps this is good to start with Softwareperson1000 (talk) 10:14, 2 June 2026 (UTC)

There's "Unix-like" as in "you can program them and use them as if they're (close enough to) Unix" and there's "Unix-like" as in "they're internally built the same way the Unixes from AT&T were". There is no requirement that a system that's sufficiently Unix-compatible from the userland programmer's point of view and from the end user's point of view be built internally the way the AT&T Unixes were built.

The kernel being interchangeable is a manifest property of it's separation from the rest of the system To what separation are you referring here? And to what are you referring when you speak of "the kernel"?

Most Unix-like systems have a monolithic kernel, possibly with loadable kernel modules. In those systems, the kernel is separated from the user-mode code by running in kernel mode rather than user mode - that concept dates back long before Unix, and most multitasking multiuser systems work that way - and have a system call interface that implements a subset of the Unix API. The kernel could be replaced, but you cannot replace one kernel with another kernel without making at least some changes to the system library the system call interfaces of the two kernels are exactly the same, from a application binary interface (ABI) point if view. The GNU C library supports platform-specific code, including the system call wrappers. This is probably due to it having been ported to other operating systems; it's not an inherent property of Unix system libraries - none of the Unix systems I worked on, such as SunOS and macOS, have C libraries structured in that fashion, their internal code design isn't oriented towards running atop multiple different kernels. This allows the GNU C library to run atop the Linux kernel and the FreeBSD kernel, for example.

GNU Hurd moves a lot of stuff that runs in kernel mode in most Unix-like systems to user-mode server processes. It does not have a Unix-like system call interface. For example, there's no single system call to do an exec call; instead, the exec calls are done with a combination of user-mode code in the library and Mach messages sent via the Mach kernel.

But what is an Unix-like design, as in is there some definition or a specific criteria for it, I don't know, I can point to you examples of several operating systems which are widely recognized as being Unix-like and I can point to some concepts of the file systems and Unix philosophy and such things None of which have anything to do with the internal design of the system.

A very wide variety of file system designs have been used as file systems for Unix - heck, Apple took a file system that wasn't designed for Unix at all and hammered it into being the primary file system for [[macOS|an operating system that even got Unix certification eventually], with a hard link implementation significantly different from what designed-for-Unix file systems have. So presumably you're talking about abstract file system properties of Unix-like systems rather than design details of particular file systems.

THe Unix philosophy is also not about internal design. Guy Harris (talk) 02:41, 3 June 2026 (UTC)

1. OK. Yet there still exists the GNU variants which are Unix-like operating systems differ in using one or the other kernel, if you don't like an "Unix" specific explanation I would give, would you then prefer editing the article to explain what the operating systems have in common versus how they differ from each other, as in the kernel and how it relates to the rest of the system, but without "Unix" specifically? As of this moment I have concluded that having such a detailed technical explanation is necessary for this article which it is currently lacking. Unix seems like the most reasonable explanation to give to describe the contents of the article, even if not unique to unix (the separation of the kernel) it is something that the kernels the GNU variants use do have. You have not convinced me that there is no need to provide an explanation in the article for what the separation of the kernels from the rest of the system, "Unix" or not I say this is necessary information to provide context for the reader who without such explanation can not be expected to know what a kernel is and thus the difference between the GNU variants (the kernel) is entirely ambiguous and unclear. Currently it just lists here are the GNU variants, and the different kernel that each use. Well what on earth is a kernel and what does that have to do with the GNU Project? The lead doesn't provide a sufficient explanation, nor the rest of the article in the current version, if I continue the recent edit war, that would provide the necessary explanation.

2. Another unrelated thing to what we were just discussing, since it's directly relevant to the conversation thread we are on as it's the topic of the original post, Aoidh apparently decided to remove the WSL from the article without consulting in the talk pages here despite there clearly being this topic here that concerns it. I understood you were in favor of having WSL in the article, would you prefer to restore the WSL back into the article, Aoidh seems to continuing on a spree to further destroy the contents of the article. Softwareperson1000 (talk) 12:04, 3 June 2026 (UTC)

Yet there still exists the GNU variants which are Unix-like operating systems differ in using one or the other kernel Yes, there are multiple systems with a userland from the GNU operating system and with kernels other than the kernel from the GNU operating system, and they are Unix-like, because the GNU operating system is a Unix-like operating system, because the GNU C library supports running atop multiple different kernels, and because the GNU userland supports running atop OSes with multiple kernels, even for the cases where kernel differences are exposed, such as different kernels supporting different ioctls.

would you then prefer editing the article to explain what the operating systems have in common versus how they differ from each other, as in the kernel and how it relates to the rest of the system, but without "Unix" specifically? Without saying that Unix-like systems all let you roll out one kernel and drop in another kernel, yes, because it's not true that Unix-like systems all let you roll out one kernel and drop in another kernel - most of them have C libraries that depend on the OS's kernel in particular.

Well what on earth is a kernel It's what the "kernel" article describes.

and what does that have to do with the GNU Project This page isn't about the GNU project as a whole, it's about the GNU variants. The GNU project included a kernel, but, as 1) the kernel part was moving slowly and 2) the GNU project's software was ported to run atop many exiting Unix-like systems, it could be ported to run atop other kernels.

I understood you were in favor of having WSL in the article WSL 1 is an example of GNU software running atop a different kernel, the NT kernel, to which had been added a bunch of changes to make it act enough like a Linux kernel for the purposes of the Linux port of the GNU C library and the rest of the GNU software in various GNU/Linux distributions (and for other non-GNU software in those distributions).

This makes WSL 1 technically similar to the Linux compatibility layers in some BSDs. Whether that renders it relevant to this article is a matter for discussion. I'm not sure I'd be firmly on the side of inclusion or removal. Guy Harris (talk) 00:12, 6 June 2026 (UTC)

I'll consider the Unix-like design and it's kernel versus userland topic settled then, unless further comments are made in the future on it I'll agree with you on this now.

Since you brought up the BSD compatibility layers as being similar, I then now propose that there could be a dedicated section for any such similar compatibility layers, which would include WSL and the BSD. Softwareperson1000 (talk) 09:30, 7 June 2026 (UTC)

I'd say that the BSD compatibility layers would belong either in Linux or Linux kernel, as it'd allow non-GNU libraries, and programs not built against any GNU libraries whatsoever, to run, and would not permit libraries built to run atop, for example, GNU Hurd to run. Yes, you could run GNU userland built with a Linux port of GNU libc, along with GNU libc itself, to run atop a Linux kernel, but you could also run non-GNU userland built with musl libc, along with musl libc itself, so it's, at most something that would support some or all of a GNU/Linux GNU variant - or some or all of a <something else>/Linux OS - to run. Guy Harris (talk) 06:01, 8 June 2026 (UTC)

I've fixed GNU variants § As operating systems to specifically note that the porting of GNU software to other operating systems was what allowed GNU variants with different kernels to be constructed; it's not as if that software will all run, with no changes, on kernels other than the GNU Hurd - there are #if/#ifdefs, multiple implementations of the same API for different operating systems, etc. involved in that. The GNU C library's source tree, for example, has non-instruction-set-dependent subdirectories of the sysdeps directory, such as sysdeps/mach/hurd and }sysdeps/unix/sysv/linux, which contain implementations of GNU APIs that run atop different kernels. Guy Harris (talk) 06:52, 8 June 2026 (UTC)

I don't quite like the way you rewrote it, I am fine with it being rewritten to make it better. However 1st now I have to question if this sentence "Software from the GNU operating system has been ported to several other Unix-like operating systems." constitutes it's own separate statement which requires it's own citation. A more important issue, it now says "other Unix-like operating systems" without it having been established earlier in the article that GNU is an Unix-like operating system in the first place, the statement merely implies GNU is an Unix-like system but this is not clearly enough communicated and will easily lead the reader to misunderstand or get confused. Also yet another issue, the current form now fails to adhere to my primary concern about providing further context to the reader about explaining what the kernel actually is in relation to the rest of the operating system. Previously before your rewrite it explained about the kernel doing the lowest level functionality and I say this should be kept, even if rewritten to be worded better. While the dedicated section above GNU Hurd explains describes the GNU Hurd kernel, it does not fully adequate for the article as the reader may skip over that segment which is specifically focusing on the Hurd and instead go straight to the As operating systems section and start reading there. It does now read that the different variants have the kernel swapped out from one to another basically and that's OK as it is certainly better than no explanation given, but it's still not the best it can be. One more thing, I think it should still read that the kernel is "taken from" one system "to form a new system". In that the GNU software is not merely ported to run on another system, which then somehow itself forms a new system from doing that, but rather the kernel is a component taken from one system by third party developers and combined with the GNU software to form an entirely new and unique operating system, which is a difference. The GNU is not merely ported to another OS, it is used in conjunction with a component from another one to form a new OS. An example, Nexenta is not GNU ported to OpenSolaris, Nexenta is the combination of taking the kernel from OpenSolaris and combining it with the GNU software to form a new unique system is a derivative of both GNU and OpenSolaris at the same time but is neither one alone. Softwareperson1000 (talk) 21:33, 8 June 2026 (UTC)

However 1st now I have to question if this sentence "Software from the GNU operating system has been ported to several other Unix-like operating systems." constitutes it's own separate statement which requires it's own citation. It's definitely true - for example, the shell in my Terminal window is bash - but a citation would be good. I'll look at adding one.

A more important issue, it now says "other Unix-like operating systems" without it having been established earlier in the article that GNU is an Unix-like operating system in the first place, the statement merely implies GNU is an Unix-like system but this is not clearly enough communicated and will easily lead the reader to misunderstand or get confused. I'll look at adding that.

Also yet another issue, the current form now fails to adhere to my primary concern about providing further context to the reader about explaining what the kernel actually is in relation to the rest of the operating system. OK, I'll look at mentioning that.

It does now read that the different variants have the kernel swapped out from one to another I guess it needs to more clearly emphasize that, in the process, some userland software also has to be adapted to the new kernel, which is the key point here. It's not as if the C library, in the form in which it's built for the GNU Hurd, can simply be used with a Linux kernel - it can't. Furthermore, any userland code that uses, for example, Mach messaging to communicate with the Mach kernel or with Hurd servers that run atop the Mach kernel will need to be changed so that, when built for Linux, it does Something Else with an equivalent effect, or just says "sorry, I can't do that on Linux".

Similarly, if the Linux kernel is replaced with some other OS's kernel, any code using Linux-specific ioctls, or Linux-specific protocol arguments to socket(), or Linux-specific socket options, or other Linux-specific mechanisms, must be changed in order to use what the new kernel provides. See, for example, libpcap, which has code to implement the same underlying APIs atop several different OSes, working around kernel and userland differences.

The good news is that, as noted, in the process of making it run atop several different OS kernels, the GNU C library had already be structured in such a way as to handle that case, so, to port it to the Linux kernel, what needed to be done was to add new system-dependent code - it didn't need to be restructured in such a way as to make it possible to do that, that had been done already.

One more thing, I think it should still read that the kernel is "taken from" one system "to form a new system". I'll look at that. Guy Harris (talk) 23:42, 8 June 2026 (UTC)

1. I've provided citations for GNU libc and GNU Bash being ported to other operating systems.
2. I've explicitly stated that the GNU operating system is a Unix-like system.
3. I've mentioned the idea of user space software and of a kernel in the lead.
4. I've improved the description of the process of combining a kernel with the user space part of the GNU operating system, modifying the user space code as necessary, to make a new system. Guy Harris (talk) 01:48, 9 June 2026 (UTC)

Linux has an RfC

Linux has an RfC for possible consensus. A discussion is taking place. If you would like to participate in the discussion, you are invited to add your comments on the discussion page. Thank you. — Preceding unsigned comment added by Victor Matheus Amaral (talk • contribs) 21:51, 5 May 2026 (UTC)

Vandalism of the article GNU variants to push a personal agenda by user Aoidh

(edited, since posting this I removed the RfC)

It is evident and clear that this article covers various operating systems derived from the use of GNU Project's software, the article being called GNU variants. GNU variants such as kFreeBSD and others, are obviously operating systems which share a commonality in their use of the software developed GNU Project, that's the subject matter of the article. It is thus absolutely relevant to this article to include an explanation of the technical nature of these operating systems in question, and their purpose for being included in this article. Their shared characteristics in their use of the GNU Project's software is the entire reason for their inclusion in this article, as well as the purpose for this article to exist in the first place. Operating systems that are considered GNU variants should be called that. With using kernels such as Hurd, Linux, kFreeBSD being the differences between the systems, thus it should be explained what causes these differences as per the Unix-like design which allows for the modularity to bring about these systems.

1 2 3

User User:Aoidh has decided on the basis of a personal point of view to claim that some of the sources that were used in the article are unreliable. And has then proceeded to vandalize the article by removing large segments of text without good reason. There was no such flaw in the sources to warrant the removal of these entire paragraphs of text when a single small edit would have been enough. Segments of text which were providing crucial context for the technical nature of the operating systems and their design. User:Aoidh in the edit notes pointed to an unrelated discussion taking place on another different Talk Page on the Linux article, on a thread which I started earlier discussing a separate and an entirely unrelated matter about "Linux" as a "family" of operating systems, which has no connection to this article whatsoever. The only supposed connection I believe User:Aoidh mistakenly saw, as to why bring that up at all here now, is that a single instance of the word "family" which I recently to this article GNU variants was used. "family" is also a key word that was the concern and the topic of that other discussion with the Linux as a "family" elsewhere. For this article, describing these operating systems as a "family" in my edit, I believe was well cited and there is no problem with using it on an article. On that other topic, User:Aoidh argued in opposition of the "family" verbiage for GNU systems and I argued in support of it, We were discussing the possibility of replacing Linux as a "family" with GNU, but this has no relation to the article GNU variants and is no basis for vandalizing it, the primary focus there on that topic was Linux which is not the topic here on GNU variants.

It would appear that as a result of that other unrelated conversation taking place elsewhere that did not concern this article, user User:Aoidh has now decided to come here to this article GNU variants to push that personal view and agenda, apparently because of the disagreement we had which has seemingly upset User:Aoidh who saw my edit on this article and thus resulted in this vandalism. It is important to realize that if the user User:Aoidh merely took issue with that single instance that word "family" being used to describe the group operating systems derived from the GNU Project's software, then the user could have simply edited out that one single sentence where that one word was used, without having to resort to vandalizing the rest of the article. User:Aoidh in the edit notes further justifies the actions taken to vandalize this article by bringin up yet another unrelated issue, the GNU/Linux naming controversy is also apparently a reason to vandalize this article. Note that this article is not concerned with the "GNU/Linux" naming controversy. As far as such disputes regarding terminology and verbiage are concerned, I will point out that this article specifically focuses on that subject matter, as per Wikipedia policy NPOV in section Due and undue weight it reads "Views held by a tiny minority should not be represented except in articles devoted to those views", this article is devoted to those views, as this article is on the topic of the GNU variants. Hence the inclusion and usage of relevant relevant verbiage concerning that topic is justified, even if it is a minority viewpoint elsewhere.

I am open to the possibility that there are better ways to explore the topic, and I now ask for comments on how that should be done. I say entirely removing the important sections of the article is not the way to go about this, I ask, if people can find better ways to explore the topic, and/or find any good better sources that User:Aoidh does not see as unreliable, that are necessary for this article.

I am opening this RfC to avoid edit warring, otherwise I would just undo the edits User:Aoidh made if I didn't suspect that would start an edit war. I am looking to restore the article back to This version, with possible further changes to be made starting from there. I request for comment on the use of the "family" term, as well as comment on the description of "GNU as an operating system" in the context of this article as that is the subject matter at hand. How should this issue be covered on this page dedicated to the GNU variants when the topic of the GNU Project's software and it's inclusion in operating systems is such a controversial issue? Softwareperson1000 (talk) 13:59, 13 May 2026 (UTC)

It seems to be a conflict between you and Aoidh. I don't think an RfC is required for this. I suggest trying WP:THIRD first. — Charles Stewart (talk) 14:52, 13 May 2026 (UTC)

I am not convinced, respectfully. Also there was already a third person involved in that other conversation User:Aoidh pointed to called User:Guy Harris who on yet another separate occasion more or less agreed with the GNU "family" idea but I am not quite sure how firm his stance is / was. My own suspicion tells me this would turn into an edit war if I didn't make an RfC. I think it's best to ask for more opinions to reach a proper consensus on this issue, for the scope of this article, as this has to do with the wider application of policy in Wikipedia as far as the Due and undue weight idea is concerned, so I'm not convinced a single third one is enough. Softwareperson1000 (talk) 15:06, 13 May 2026 (UTC)

But the more important reason for an RfC is that User:Aoidh is the one who brought up another unrelated "consensus" on the topic of Linux and is also the one to raise the unrelated GNU/Linux naming controversy which has no relation to this article as a justification for the acts. Secondly I don't see that there is anything between me and User:Aoidh in particular as you would put it. The other discussion is entirely disconnected from this article, so in my mind that other discussion is entirely unrelated and I don't care about it now at all now. User:Aoidh is the one who brought it up where it doesn't belong. But I think that it does now warrant an RfC for the scope of this article since the questions regarding such controversy has been now introduced here, as it is a controversial issue. Softwareperson1000 (talk) 15:23, 13 May 2026 (UTC)

Again, please stop making WP:ASPERSIONS without evidence. I removed the recent additions because of several issues such as unreliable sources and failed verification, as well as WP:OR and longstanding consensus; this discussion is relevant because it is the same argument at a new article when consensus did not support the content there, and coming to this article to make it a WP:POVFORK is an issue. I did not revert these recent changes on the basis of a personal point of view that some of the sources are unreliable; this is not a reliable source per WP:SPS, for example. Conflating Linux (and the less common GNU/Linux label) with GNU itself and using that basis to conclude in the prose things about GNU itself based on sources for Linux is a violation of the Wikipedia:No original research policy. There is nothing about my reverting based on these issues that would make it vandalism (WP:VANDTYPES), and accusing me of trying to push a personal agenda without evidence is unnecessary, and I respectfully have no interest in engaging with comments of this nature any further. - Aoidh (talk) 17:33, 13 May 2026 (UTC)

OK so the p2pfoundation one is bad then, so remove that one citation. That wasn't the only citation for the sentence it was used for however as there was also this one https://www.getgnulinux.org/linux/, so I take it that was no good either? On the part about GNU as a family I admit there were too many citations used, however I do have a reason for that, I have added the numerous citations to the article with every intention of eventually removing many of them which are bad later, and/or letting other editors to review which sources are good and remove the bad ones. I have taken this article as a hobby project, where I contribute to it incrementally, I consider this article an active work in progress where I have added the multiple citations there as notes for me for further scrutiny later, they should be good enough to be included in the article but it's true there are too many. I realize this is not good practice would probably violate some policy, but my intention was to get back to them soon enough. And you will see evidence of me already doing this practice recently.

That issue, nor the one citation you said is bad, does not alter the fact that This citation is all good, also I would recommend This one as well as good, these ones I don't see as being bad, the other ones for the "family" issue are probably not useful so they can be removed. But don't pretend as if that one p2pfoundation link was all there was as a citation to warrant the removal of so much more other text. What about the part about the Unix-like design which is the whole reason for why the multiple kernel variants exist? I say it makes sense to explain why the different kernel variants exist on an article that covers those systems, was it not well cited? Or what about this part "GNU distributions are available for a wide variety of computer systems ranging from..." Which predates my edits to this article, why remove it? It's just a generic fact about things that are obvious from the context of the rest of the article, not literally every single line of text has to be cited as per Wikipedia policy, sometimes the rest of the article can just give the rest of the context for such a generic statement.

You did vandalize the article, because as it stands it is a mess and is incoherent, for some reason after your edit, GNU Hurd now has two different sections back-to-back which is nonsense, and there is no clear separation between the complete OS variants which was the GNU as an operating system you removed, versus the other sections for GNU distributions for use on other operating systems and Impact which were categorized as such under the context of separating GNU as an OS versus those other forms of GNU variants, showing that you did not think this trough / have an intent to make the article actually make sense, but instead what you did was showed up here to remove a bunch of stuff arbitrarily that you didn't like and left it in that sorry state without thinking of the article as a whole for how these changes fit in.

Also since you so seemingly so strongly support the case that GNU isn't an OS of any kind (never mind the family thing) why did you leave "...as well as to form complete operating system distributions out of..." then? That literally says GNU variants are still complete operating systems? If you are going to push your personal viewpoint then might as well eradicate any mention of the term "operating system" in this article if GNU isn't that? This suggests you haven't though this trough or carefully analyzed the article to understand it, and instead took a quick glance at it and didn't like the part where it said "family" and decided to just leave that part, this is why I say what you did is vandalism because that's what your actions certainly look like. As to your personal motivations for why and what brought this about, those are my speculation, vandalism is my argument.

I will repeat, that the discussion about the "Linux" family is irrelevant. You will note, again for the millionth time, that in that other discussion I said I am OK with the "Unix-like" as being the family as a compromise, which literally means that conversation wasn't about the GNU Project specifically, otherwise if it was I would have insisted it has to be GNU and nothing but GNU, the fact that I offered other solutions as a compromise shows GNU was not the primary focus of that discussion, it was my primary recommendation for the replacement of "Linux" but it wasn't the topic of that conversation. Also, an important note is that the scope of that discussion was different, I made the case as we spoke that all the articles on Wikipedia should reflect that. That is fundamentally a different context than, an article where the subject matter specifically focuses on the GNU system variants A.K.A. GNU as an operating system, as being described as a "family". That's different, the context and scope change the whole discussion.

But OK, let's drop the "family", I'll agree to that as a compromise and I'll restore the article to what it was before your edits then? Softwareperson1000 (talk) 12:12, 14 May 2026 (UTC)

P.S. There is currently no link to the article GNU because you removed it and didn't clean up / fix the mess you caused, which means you have contributed to an overall worsening of the connectivity between articles in Wikipedia and disconnected relevant pages from this article. A total mess is what you left behind. Softwareperson1000 (talk) 12:36, 14 May 2026 (UTC)

As I said, I'm not going to respond to comments like this. Get a consensus for your changes before trying to reinsert the material further. - Aoidh (talk) 22:10, 14 May 2026 (UTC)

This is not a personal attack however, it is a valid case to point out vandalism done to the article. I have provided a motive, an established personal viewpoint, the problems caused to the article, as well as clearly explained the series of events. You can not evade personal responsibility and accountability by naming any attempt to call out such behavior a "personal attack" to attempt to dismiss it on such grounds. You have vandalized the article.

Section title "GNU as an operating system" which points to GNU variants as the 'Main article' in article GNU is my consensus. And I will restore the article to what it was before you vandalized if I so see fit. You have no defense, you have pointed to policies and called things names.

Let me tell you something about secondary sources, secondary sources are needed when the primary sources are patents which are complex legal documents that are not written in plain language such that a lay person would understand it, thus citing an expert's review of that patent as a secondary source is necessary for proper citation. But we are not dealing with patents here (or equivalents).

Since you like pointing to Wikipedia policy so much, let me show you how it's done: In  Wikipedia policy No original research Section Primary, secondary and tertiary sources link here it reads "Appropriate sourcing can be a complicated issue, and these are general rules. Deciding whether primary, secondary, or tertiary sources are appropriate in any given instance is a matter of good editorial judgment and common sense, and should be discussed on article talk pages."

Your claims that secondary sources are a hard requirement are false and your case is thus dismissed on these grounds.

Let me tell you something about common sense, if you open your eyes you will see the clear and obvious facts, 1. you will see there are lots of operating systems which make use of software derived from the GNU Project 2. The operating systems are historically derived from the GNU Project specifically meaning that Richard Stallman started the project and made some software, and then third parties took that software and threw in some third party software to form a complete operating system 3. This article obviously covers said operating systems 3. They are called GNU variants 4. they are also called GNU distros

Then with those obvious facts in mind, here is my question for you: What on earth do you think an operating system that is considered GNU variant is, if not an operating system that is considered a GNU variant? Softwareperson1000 (talk) 10:02, 15 May 2026 (UTC)

I'm happy to discuss the issues, but I'm not going to engage with WP:ASPERSIONS further. - Aoidh (talk) 19:18, 15 May 2026 (UTC)

Let's say we did drop the "vandalism" then, I expect you to then answer the questions I have asked you thus far if you are happy to discuss the issue. Mainly, what do you think a "GNU variant" in the context of a complete operating system distribution is, if not a variant of the GNU operating system. I am fairly certain that Richard Stallman was the one who coined the usage of "variant" in the context of GNU system distribution, specifically describing Linux using operating systems which use the GNU software as being each a modified variant of GNU with Linux added essentially. And I believe it is this terminology that is used on this Wikipedia article is derived from that use of GNU as a "variant". Which means, already implicit in the use of "GNU variant" is the acknowledgement that GNU manifests as an operating system in the form of this and that system distribution, because that's what it means, GNU variant.

So how can you continue your support (in not actively opposing it) of "GNU variant" but then you refuse to acknowledge "GNU as an operating system"? That's a bit silly now isn't it? Because the two are equivalent and GNU variant stands for GNU as an operating system in that original context Richard Stallman used it in.

Then the second question is, how come you removed so much text from the article if it wasn't arbitrary vandalism? You have pointed to a single citation which wasn't to your liking, but there was a lot of other text you removed. Also none of it had anything to do with any GNU/Linux naming controversy which you for some reason brought up as if it has anything to do with this.

Then how do you explain why you ruined the article, by making such a mess of it as I explained earlier?

What's with you not having edited the section title on the GNU page in the last 6 years?

You have no valid case against the use of "family" anyway, as any consensus on the GNU/Linux naming controversy you think you are citing, has nothing to do with this, and I have not seen a single link or a quotation to any specific statement disputing "family" in such a context. Not once from you, so you are just making stuff up.

I would appreciate addressing the argument I made about the use of primary, secondary, and tertiary sources that has to do with editorial judgment and such things.

And yes, Wikipedia does have a specific meaning for Vandalism, and it opens with "The malicious removal of encyclopedic content, or the changing of such content beyond all recognition, without any regard to our core content policies of neutral point of view (which does not mean no point of view), verifiability and no original research, is a deliberate attempt to damage Wikipedia. There are, of course, more juvenile forms of vandalism, such as adding irrelevant obscenities or crude humor to a page, illegitimately blanking pages, and inserting obvious nonsense into a page. Abusive creation or usage of user accounts and temporary accounts may also constitute vandalism."

Later in Types of vandalism it reads Vandalism on Wikipedia usually falls into one or more of these categories: which essentially means included but not limited to

Does it not qualify what was done, as a malicious removal of content and without regard to the Neutral Point Of View to push a personal viewpoint, and I have presented the motive, that established personal viewpoint, the damage done to the article, and explained specifically what happened and how things played out? And this here is a question not an accusation.

Why does accusing of vandalism stop the ability to answer the questions asked? How does this differ from if I did stop the accusation, as in, what changes, the questions are the same and the issue remains, would you then answer the questions you have refused to answer thus far if I no longer accuse you of vandalism? Softwareperson1000 (talk) 11:59, 16 May 2026 (UTC)

correction "opens with" is incorrect, it is the second paragraph Softwareperson1000 (talk) 12:26, 16 May 2026 (UTC)

No one is required to satisfy you with their answers, especially when the excessively long comments are repeating the same aspersions over and over. - Aoidh (talk) 19:28, 16 May 2026 (UTC)

You need to stop accusing Aoidh of vandalism with no basis. Your dispute of the content of their edits has nothing to do with vandalism, which has a particular meaning on Wikipedia. Eyesinthefire (talk) 20:23, 15 May 2026 (UTC)

But there is basis. The case I have made is valid. I am open to the possibility I am wrong, and I am willing to dismiss the case if so determined, I am open to hearing counter arguments. On the particular meaning of vandalism, yes, and the most important part of it is this one "The malicious removal of encyclopedic content, or the changing of such content beyond all recognition, without any regard to our core content policies of neutral point of view (which does not mean no point of view), verifiability and no original research, is a deliberate attempt to damage Wikipedia." Softwareperson1000 (talk) 12:24, 16 May 2026 (UTC)

P.P.S Are you also going to undo the edit made here in 4 August 2020 to the GNU article, which introduces the "GNU as an operating system" to that article "Per talkpage." it says in edit notes? A line of text used as the heading for that section established almost 6 years ago now to say so, which equates the GNU variants with GNU as an operating system. You will see, that section in that article points to this article GNU variants as the 'Main article', hence GNU variants distributed as complete operating systems are that, GNU as an operating system. It looks like to me, that you User:Aoidh are the one countering and evading established use of terminology to refer to the systems, to push your own personal view. I can see in the Talk page archives, you have been there as part of the discussions, you have a long established personal anti-GNU agenda that is evident and clear. My question for you now is, why haven't you in the last 6 years undone that edit made there to the GNU article, and it was only now when I talked to you and when I made these changes to the article GNU variants, that you came here to this article GNU variants to undo parts that refer to the GNU variants as an "operating system" specifically, but have not cared about that part on that GNU article? A line, GNU as an operating system, that you have very clearly long argued against, personal viewpoint. So, did you think you could get away pushing your own personal narrative on another separate article now, so as to not counter that established line of terminology used there by raising that old issue there again? That's speculation on my part if this was your motivation, but it certainly does not add support to your case that this was not vandalism and your personal view and mission, what is clearly a well established personal view on your part. You say I have no evidence you are pushing a personal agenda, is this not it? Softwareperson1000 (talk) 13:58, 14 May 2026 (UTC)

@Softwareperson1000: what is your brief and neutral statement? At over 6,000 bytes, the statement above (from the {{rfc}} tag to the next timestamp) is far too long for Legobot (talk · contribs) to handle, and so it is not being shown correctly at Wikipedia:Requests for comment/Wikipedia policies and guidelines. Also, per WP:RFCCAT, The "Wikipedia policies and guidelines" category is for discussing changes to the policies and guidelines themselves, not for discussing how to apply them to a specific case. --Redrose64 🌹 (talk) 21:15, 13 May 2026 (UTC)

Also there was already a third person involved in that other conversation User:Aoidh pointed to called User:Guy Harris Said third person is called that because Mr. and Mrs. Harris decided to name their son "Guy" and because said son is one of those weirdos who just uses his legal name when posting.

who on yet another separate occasion more or less agreed with the GNU "family" idea I'm not sure to which comment you're referring - perhaps the comment that begins with "For Linux as a family: the references seem good. " on Talk:Linux (I'd have linked directly, but Wikipedia can't seem to handle the permalink URL that it generated if I put it in the comment) - but I'd say, based on what the GNU folks say, that they, at least, speak of such a family, with "the GNU system" being what most people might call "GNU/Hurd" and GNU/Linux being a "variant" of the GNU system. I'm not sure how they define the non-kernel part of the GNU system, or how much of it must be present in a system for it to be classified as a GNU variant. The system on which I'm typing this has bash as a shell (although it switched from bash to zsh as the default, and /bin/sh is a wrapper that runs a shell that the user can specify) and GNU make as its make, and that may still use other GNU userland tools even though the vendor has made an effort to replace GPLed tools with non-GPLed tools over time, but I suspect the GNU folks would be offended if you called it a GNU variant. :-)

but I am not quite sure how firm his stance is / was The tl;dr version of my stance is "it's complicated", because it is complicated. (Some might wish it to be uncomplicated, but if wishes were horses, beggars would ride.)

But thanks for pointing out that this page is another page that weighs in on this issue. Guy Harris (talk) 21:31, 13 May 2026 (UTC)

Sorry, I mistook you for someone else. My memory deceived me, I should have verified this before I mentioned you agreed, my mistake. It was some anonymous account on the same thread I talked to you on on the initial GNU family proposal thread. This is what I falsely recalled Softwareperson1000 (talk) 11:25, 14 May 2026 (UTC)

RFC:This isn't a properly formulated RFC and requesting comments at this stage is premature. I would suggest actually working through this first. Wikipedia:BEFORERFC Lukewarmbeer (talk) 11:40, 14 May 2026 (UTC)

Fair enough, after having heard such criticism also earlier that RfC isn't the way to go now, I have decided to drop it. Softwareperson1000 (talk) 11:53, 14 May 2026 (UTC)

Also, I want to point just in case there was some misunderstanding, that I did not intent to insult your name User:Guy Harris. When I said you were "called" that, I literally just meant that in the most plain generic way possible possible, where that's just literally what the user account is called. So there was no intended implication about the quality or characteristics of the name, just so we are clear I have no issue with your name. Softwareperson1000 (talk) 13:06, 14 May 2026 (UTC)

No insult inferred, it was just odd phrasing. Relatively few people seem to just use their legal name for their Wikipedia accounts, so I just wanted to clarify that it's not a handle, it's me. Guy Harris (talk) 19:55, 14 May 2026 (UTC)

It's also probably worth pointing out to avoid further confusion with there was already a third person involved that User:Guy Harris and User:Guy Macon are different people. - Aoidh (talk) 22:20, 14 May 2026 (UTC)

And I also use my legal name. Feel free to call me "that Slackware Guy". I am guessing that there is only one of those... (smile). --Guy Macon (talk) 23:20, 14 May 2026 (UTC)

References

Sourcing

There are sixteen references in the single-paragraph lead. There are 12 references in the last sentence of 14 words. While there is no specific mandate, content in the lead is considered summary information that should be found (and sourced) in the body of the article, with some exceptions. Certainly, there is no need to add sources beyond what is needed to verify information. -- Otr500 (talk) 13:54, 15 May 2026 (UTC)

I agree it's unnecessary, and I've restored the version of the lede prior to the recent edits because of the WP:OVERKILL issue as well as other issues introduced by these edits, such as MOS:LEDESENTENCE. This version does not follow Wikipedia's Manual of Style nor does it provide proper attribution to who is calling it a variant. - Aoidh (talk) 18:55, 15 May 2026 (UTC)

I agree with the above criticism of the rewritten lead, but I want to add that between the lengthy parenthetical mid-sentence and the extraneous dependent clauses, I found the opening sentence incomprehensible. The opening of an article is supposed to be clear and direct, giving a quick summary of what a reader needs right up front. The new lead was far from that ideal, so we are better off sticking with the older one. MrOllie (talk) 18:01, 18 May 2026 (UTC)

This version https://en.wikipedia.org/w/index.php?title=GNU_variants&oldid=1354867000 fixes the problem with too many sources. I admit it was a mistake on my part in having too many of the sources, I explained in another conversation thread the intention of my actions Talk:GNU variants#c-Softwareperson1000-20260514121200-Aoidh-20260513173300 and I believe this issue has been resolved in that version of the article I prefer which I pointed to above.

As far as the lead being incomprehensible, this version I point to is not that, it is a very clear and concise way to put it. The issue is, that the GNU variants as a topic, as explored by the article, are perhaps difficult to understand, sometimes GNU variants are complete system distributions, and other times software for use on another system as with the GNU-Darwin CD distribution for use on a Darwin installation as well as Cygwin's software distribution including GNU software targeting Windows NT. If the topic itself is naturally complicated, then that is an issue with comprehension of the subject matter at hand, not an issue with the wording of the explanation of the subject matter. I believe there is no fault in that version in the above link to the version of the article I prefer. To simplify it to make it easier to understand would be oversimplifying it to the point that it actually would fail to accurately describe the article. As it stands, the lead of the article does not account for Cygwin at all, which means either the current lead is inadequate, or Cygwin should be removed from this article, as Cygwin is not an operating system! (exclamation mark for emphasis of importance) Softwareperson1000 (talk) 11:16, 31 May 2026 (UTC)

RfC: Resolve Edit War between me and User:Aoidh

Resolve what to with the article after the edit war started when User:Aoidh recently entered the scene. See https://en.wikipedia.org/wiki/Talk:GNU_variants#c-Softwareperson1000-20260513135900-Vandalism_of_the_article_GNU_variants_to_push_a_personal_agenda_by_user_Aoidh for context

Question A

1. do you agree with my support for restoring the article to look like this https://en.wikipedia.org/w/index.php?title=GNU_variants&oldid=1354867000 as the base version for future improvement or the version preferred by User:Aoidh 2. Do you agree with using the version preferred by User:Aoidh instead https://en.wikipedia.org/w/index.php?title=GNU_variants&oldid=1354487284 3. something else? explain

Question B

I introduced in the edit https://en.wikipedia.org/w/index.php?title=GNU_variants&oldid=1352197669 a new system of categorizing the different entries on this article GNU variants into 4 distinct categories the categories: GNU as an operating system GNU distributions for use on other operating systems GNU software used in other operating systems Impact

The system of categories I introduced was tampered with by User:Aoidh and sustained efforts from User:Aoidh as contributing to the edit war prevented me from restoring the "GNU as an operating system" section title that was removed. In edit https://en.wikipedia.org/w/index.php?title=GNU_variants&oldid=1354267249 I offered as a compromise to retitle "GNU as an operating system" as "Complete system distributions" to avoid terminology calling GNU an operating system which apparently was the argument User:Aoidh made, which is unclear as User:Aoidh to this date has not made a comprehensive explanation

1. entirely undo the separation into 4 categories I originally introduced 2. restore the 4 categories to either how I originally put them. Alternatively retitling "GNU as an operating system" as "Complete system distributions" which I accept 3. use the version Preferred by User:Aoidh, the categorization using 3 of the 4 categories I introduced but with "GNU as an operating system" or alternatively "Complete system distributions" removed, which breaks the context for why I introduced the 4 categories. 4. You have something else as a better suggestion? explain

Question C

Do you agree with me when I ask to include an explanation of Unix-like systems design as the necessary context for explaining to the reader what differentiates the GNU variants from each other, that being the kernel, such as to explain what the kernel is in the first place and importantly how that relates to Unix-like operating systems which are a shared commonality of the GNU variants on the article? Given this article covers Unix-like operating systems with the predominant differentiating feature between the various GNU variants being the "swapping out" of one kernel for another, explaining that relation between the kernel and the rest of the Unix-like operating system as a key detail is crushial to provide the reader of the article context for what is going on. Without this explanation of Unix-like design, the GNU variants on their own as mere entries on this article, are too ambiguous and fail at establishing what makes a system a GNU variant, and how they differ. Exploring changing one kernel for another, is enough of a technical matter to warrant such an article to explain to a reader who might stumble upon this article and first learns about kernels here what they are and how they relate to the operating system.

Question D

User:Aoidh removed several, arbitrary segments of text which have no explanation given for the reason for removal. In the earlier edit notes before the edit war User:Aoidh points to an unrelated discussion having to do with the use of the term "family" to refer to operating systems, as well as the GNU/Linux naming controvercy, as the main reason to intervene in the article, the question, what to do with each other segment of text that was removed but does not mention the words "family" or "GNU/Linux" with no further explanation given for the reason for removal of such segments of text?

Question E

Do you agree with me when I say the the sourcing is lower in quality in the version preferred by User:Aoidh and is a regression as compared to the improved sourcing in my version https://en.wikipedia.org/w/index.php?title=GNU_variants&oldid=1354867000, where for example User:Aoidh does not make use of archive.org links which I do use, but more importantly one of the sources User:Aoidh prefers which is used in the first sentence of the lead paragraph literally does not feature the word "GNU" anywhere in the source? This one https://www.makeuseof.com/tag/6-lightweight-linux-distributions-give-pc-lease-life/ where who knows if it maybe once did, years and years ago (no archive.org links used by User:Aoidh stated previously), it seems User:Aoidh did not actually bother to read the sources that were used, displaying a total lack of competence in selecting sources by User:Aoidh. Several of those other sources are also trash such as this one http://en.linuxreviews.org/The_GNU_Operating_System and also this one https://www.outreachy.org/2018-may-august/communities/gnu-guix/ which hardly contribute anything of any meaningful substance to the article, overall absolutely terrible sources used by User:Aoidh compared to my preferred version.

Question F

I hold that this is vandalism done on the part of User:Aoidh as my personal opinion, where User:Aoidh essentially showed up and ruined the article, making a terrible mess out of it, do you agree or disagree?

Answers

Softwareperson1000 (talk) 11:34, 31 May 2026 (UTC)

For the benefit of others, I'll point out that I'm aware of this RfC but will not comment beyond this. This isn't the first RfC created to cast aspersions and focus on conduct as much as content, and I am not the only one to revert the editor or to disagree with them on this talk page about the changes made. I'm happy to explain the issues with the edits in as much detail as needed if and when the focus is on the content rather than repeated aspersions about myself and my apparent "total lack of competence". - Aoidh (talk) 14:50, 31 May 2026 (UTC)

"I am not the only one to revert the editor"

yeah, the other one undid my RfC Softwareperson1000 (talk) 11:57, 1 June 2026 (UTC)

• BAD RFC. Bad enough that I removed the RFC tag. RFCs are supposed to be a brief, neutral statement, not a long argument for one side of the issue sprinkled with personal attacks on another editor. - MrOllie (talk) 17:04, 31 May 2026 (UTC)

  You requested a consensus in your edit notes https://en.wikipedia.org/w/index.php?title=GNU_variants&oldid=1354867685 if you don't like it when I start it, then why don't you start one? Softwareperson1000 (talk) 11:54, 1 June 2026 (UTC)

I am only here for the RFC. I do not see this topic, or the page in general, as being in any form that I can comment on helpfully. This standard of slanging match contributes nothing to WP, or to the usability of the RFC tool. Either drop the topic or appeal to the disciplinary authorities. JonRichfield (talk) 18:10, 31 May 2026 (UTC)

correction, while the concept of separating the article into the distinct categories was indeed introduced in this edit https://en.wikipedia.org/w/index.php?title=GNU_variants&oldid=1352197669 the number of those categories was in fact not yet 4 but 3 initially, it was only until edit https://en.wikipedia.org/w/index.php?title=GNU_variants&oldid=1352974180 where the 4th category was introduced "Impact". Which does mean that by the time Aoidh entered the scene it was the 4 categories and it still would be in the version I would prefer in the edit wars. Softwareperson1000 (talk) 12:52, 2 June 2026 (UTC)

OpenSolaris and Illumos kernel

This article says:

Nexenta OS is the first distribution that combines the GNU userland (with the exception of libc; OpenSolaris' libc is used) and Debian's packaging and organisation with the OpenSolaris kernel. Nexenta OS is available for IA-32 and x86-64 based systems. Nexenta Systems initiated the project and sponsors its continued development.

However the Nexenta OS article says: "In late 2011, the Nexenta OS brand was terminated and replaced with Illumian" so there is a need for some clarification. Has the project been revived? Has a different project adopted the name? --Northernhenge (talk) 15:15, 20 June 2026 (UTC)