Talk:GNU/Linux naming controversy/Archive 4

From Wikipedia, the free encyclopedia

 This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
← Archive 3 Archive 4 Archive 5 →

Contents

lack of legal enforcement is not an argument for one side or the other

Adrian keeps inserting the following under "Arguments for Linux"

"No legal requirements or trademark restrictions to use the terms "GNU", "Linux" or "GNU/Linux" are stipulated by the individual software packages or an aggregate distribution. People that name a product have the freedom of choice to use "Linux", "GNU/Linux" or not even using "Linux" at all in naming their products that are based on GNU and Linux projects. Most of the developers and companies use the term Linux to describe the Operating System that is based on Linux kernel and GNU tools, one notable exception is "Debian GNU/Linux".

I have removed it, for three reasons:

  • First, the fact that people are legally free to choose either name is, with a moment's thought, not an argument per se for one name or the other.
  • OK, but it has to be mentioned in order to frame the discussion. People are free to call their products however they want, nobody can tell them how to call them even if they are based only on Linux or GNU code, actually the only requirement is to not call them Linux or GNU unless they get permission for people who handle the registred names. -- AdrianTM 02:13, 6 August 2006 (UTC)
  • Second, the fact that people have, de facto, mostly chosen a particular name has already been mentioned several times (pointing out that e.g. the "Linux" name is the most popular, the oldest, and has the most historical momentum).
  • Third, even if you feel you have some further smashing point of logic to add, we've being trying hard to avoid arguments by Wikipedians here, following policy. As much as possible arguments should be attributed to prominent commentators and supported by citations. (Several of whom have already been quoted as pointing to the popularity of the "Linux" term in response to suggestions of "GNU/Linux".)

—Steven G. Johnson 14:06, 29 July 2006 (UTC)

cleaning up the intro

the intro felt rather POV-ish to me, consisting mostly of claims in favor of GNU/Linux proponents. i've shortened it and tried to make it more objective; those same claims can already be found in the section explicitly devoted to them. Benwing 02:38, 3 August 2006 (UTC)

it seems more POV-ish to me and less objective. where is the source for "GNU-written software is only a fraction (often cited as 25-35%) of the code that forms the Linux operating system"? --71.161.215.74 04:01, 6 August 2006 (UTC)

I partially agree with this, people, everytime you come up with some numbers or estimates please, please, provide the reference, it shouldn't be dificult, just use the <ref> </ref> tags and include the link or the title of the book that you got the info from. Thanks. -- AdrianTM 04:16, 6 August 2006 (UTC)
Besides the fact that it is a straw man. If you read the article, you'll find that the question of the size of GNU's contribution (although it is certainly larger than the kernel) to a typical distro is not the FSF's main argument. However, if you want numbers the "Linux and the GNU project" by Stallman cites "one CD vendor" as estimating 28% of their distro came from GNU. David Wheeler's "More than a Gigabuck: estimating GNU/Linux's size" article [1] unfortunately doesn't give number, although it states that "the total of the GNU project's code is much larger than the Linux kernel's size" while the kernel is the "largest single component". Unfortunately, no recent, quantitative sources that I can find in a quick search, but as I said it's a straw man anyway. —Steven G. Johnson 14:21, 6 August 2006 (UTC)
I don't agree it's a straw man, it's a relevant fact. It doesn't argue against RMS arguments to be a straw man it's an argument in itself: "hey look this is the size of GNU, this is the size of Linux and this is the size of other parts", now you could argue that the size argument is not important (although we all know "size do matter") but you can't say it's a straw man. Frankly anytime I see somebody complaining about straw men in a discussion I know they are out of arguments... -- AdrianTM 16:38, 6 August 2006 (UTC)
Um, what part of "don't put words in people's mouths" do you not understand? The intro presents this as the main argument of proponents of the term. It is not, as can be easily verified by RTFA. I agree that the size of the GNU components and the size of the kernel are relevant facts, but they should be presented as such, not as a false basis for an argument. I've rephrased the intro to try and present this more neutrally and accurately. —Steven G. Johnson 19:13, 6 August 2006 (UTC)
I'm not sure you understand, they make a point in saying that the size of GNU code is small, that's a valid point in itself not a straw man. The issue is not what "GNU/Linux" or "Linux" proponents said, the issue is what the reality is, the GNU code is 20-30% of Linux distribution and that's an argument in itself. I can understand to correct the first paragraph that says that "Proponents of the term claim that the Linux kernel is only a small part of the system as a whole, and that the rest of the system is dominated by GNU software" if that's not correct, however I don't understand why you should delete that GNU is only 20-30% of the Linux OS, which is a valid point since more than 70% is something else, maybe Linux should be called something like Linux/GNU/BSD/Trolltech/etc OS. As far as I understand that should be only an introductory paragraph that shows what the discusion is about and arguments should be presented under the appropriate headings, however as it stands it seems like it presents a POV rather than just presenting the discussion. -- AdrianTM 20:24, 6 August 2006 (UTC)
As a side remark what FSF claims that "Linux kernel is only a part that was missing from the GNU system" is pretty silly, here Linus is right, if they would take Linux kernel and add it to GNU they get to call that whatever they want, even simply "GNU" or "GNU/Linux" if they want to be pedantic, but since Linus took the GNU code and added to his kernel he gets to call the final product however he wants: "Linux", "Freax", etc. Same for distro maintainers, they get to call their products whatever they want. All this discussion is pretty useless because things are called how people call them, not how they "should" be called. Ask any linguist... -- AdrianTM 20:24, 6 August 2006 (UTC)

Adrian, the question is not whether something is "an argument", but whether it is an argument that prominent commentators actually make, as supported by citations. The tendency of this issue towards endless flamewars is a good reason to stick to closely to actual quotations. Second, the reason it was a straw man is that the size of the GNU system was falsely presented (in the old intro) as the main argument of the FSF and supporters, with the "but GNU is a small portion" as a response to this. Why is this so hard to grasp? Present the basic facts, and the arguments actually made.

(By the way, let's try to avoid hashing out our own arguments for the name here. Our personal opinions on this issue aren't particularly relevant to the article.)

—Steven G. Johnson 21:21, 6 August 2006 (UTC)

I didn't add that section about 20-30% GNU code in Linux OS, that's why I asked for reference too. I don't claim though that it wasn't said by some "prominent commentor" since I don't see any reference and I don't know who said it. -- AdrianTM 21:28, 6 August 2006 (UTC)
So what it is your objection? If it's not by a prominent commentator and is unsourced, it should be removed, and it was. Anything put in the introduction is implicitly a summary of the sourced material later in the article, and therefore is implicitly a summary of views of prominent commentators. —Steven G. Johnson 21:33, 6 August 2006 (UTC)
My objection was that it's not a straw man as you claimed. I do agree with you that's not referenced. sometimes instead of removing stuff it's better to add a {{fact}} or {{verify source}} tag. Your edits seemed to me a little bit lopsided, but that's probably OK, if somebody comes with the reference they can re-add it in the right place. -- AdrianTM 21:40, 6 August 2006 (UTC)
The topic may be intrinsically lopsided, I'm afraid. I searched very hard for quotes to put in the "Arguments for Linux" section. (IIRC, all of the quotes there were inserted by me.) Most prominent commentators who use the "Linux" name simply ignore the debate—they don't have to argue since the status quo is their position. There already are a couple of quotes (see the Gettys quote, for example) that I added, however, which do argue against GNU/Linux on the basis of there being many contributors to the system. —Steven G. Johnson 22:13, 6 August 2006 (UTC)
Regarding the intro, though, there are now two sentences laying out some basic facts (what is a Linux system and what part of it is the kernel and what part is GNU components, and other stuff...I'd love to include the 30% number if a reputable source can be found, although to be fair you'd want to also quote a number for the kernel). It has one sentence summarizing the FSF position, and one sentence summarizing the arguments on the other side. I'm not sure how this is "lopsided". —Steven G. Johnson 22:13, 6 August 2006 (UTC)

Steven, I was the one who rewrote the intro that you rewrote again; and I've rewritten it yet again. You seem to have misunderstood some of my intention, so I've clarified it with further hedge terms and such. 20-30% is *not* an objective figure (there is no such figure, since there's no single definition of what a Linux OS is); rather, it's part of a claim made by proponents of using just "Linux". It's *not* a straw man; in fact, it's one of the most common claims made on this side. Your arguments that size "doesn't matter" in the FSF's view are clearly not quite right, since RMS's claims that the kernel is the "final small part" (or whatever) of the GNU system clearly imply that GNU is all the rest. However, I see your point. The sticking point seems to be in the validity and importance of the "GNU system" as a concept attributable to RMS, vs. code actually written by the FSF. I added a paragraph on the difference between the GNU system and actual GNU-written software that tries to clarify this.

I removed your text about size of the kernel itself vs. all GNU-written software; I don't think it's useful and it rather confuses the issue.

I also took out some text about trademarks because I don't think it's relevant in this context; but here it is:

(No attempts have been made to legally enforce a particular name, through trademarks, software licenses, or other means, although the term "Linux" is trademarked.[1] For example, no particular name is required by the GNU General Public License, used for both the kernel and most GNU code.)

Benwing 09:12, 7 August 2006 (UTC)

If you chose to remove "text about size of the kernel itself vs. all GNU-written software", because you "don't think it's useful and it rather confuses the issue", then it's a bit contradictory or unfair to have left this in the intro, "Proponents of the term 'Linux' assert that the term 'GNU/Linux' exaggerates the importance of software that is claimed to comprise only a fraction of the code that forms the Linux operating system as a whole". --69.54.29.23 20:52, 18 August 2006 (UTC)

No attempts have been made to legally enforce a particular name

This misrepresents the issue. It's not that there was no attempt (yet) the issue is that is part of the freedom that GPL gives you, there's absolutely no requirement to name the derived code in a way or another. (correct me if I'm wrong) -- AdrianTM 20:37, 6 August 2006 (UTC)

Actually the only legal issue is not to use a trademark if you don't have the permission. But that's beyond this discussion. -- AdrianTM 20:39, 6 August 2006 (UTC)

To say that the GPL means that there can be no restriction on naming is a misunderstanding. First, the GPL is orthogonal to trademark law. Since "Linux" is trademarked, the Linux Mark Institute could decide tomorrow to prevent people (or at least products) from using the "Linux" name for anything but the kernel proper. Second, since the FSF owns the copyright on most of the GNU code, it could freely change the license tomorrow to require any operating system distributed with GNU code to be called "GNU/something". Third, I'm sure a lawyer could come up with other tricks.  :-)

I'm not suggesting that either of these things is likely to occur. But it is false to imply, as the article did, that it is impossible for anyone to try and legally affect the nomenclature because of the GPL. The only correct thing is to say that this has not happened.

—Steven G. Johnson 21:03, 6 August 2006 (UTC)

We can't speculate what FSF might do or not, right now GPL doesn't require name attribution. Anybody can use the code and can do whatever they want with it, including making products that are named something else than "GNU". That's the fact. -- AdrianTM 21:10, 6 August 2006 (UTC)
I really don't understand your problem here. The article states that no one has attempted to enforce a particular nomenclature, which is a fact that we can apparently both agree on. The old version was misleading because it implied that legal measures are impossible, which is clearly not true as I explained. Also, it is false to claim that there are no legal restrictions whatsoever on the naming, because the Linux Mark Institute does enforce the "Linux" trademark and does impose conditions (e.g. it charges a fee in certain contexts). What do you want it to say? —Steven G. Johnson 21:27, 6 August 2006 (UTC)
the legal measures are impossible as for now there's no requirements in GPL for naming of the derived products, we can't speculate about the future, but even GPL v3 doesn't contain such clauses (which by the way would be against the principles of free software). How is it now, I made it clear what are the restrictions for Linux name? -- AdrianTM 21:31, 6 August 2006 (UTC)
I liked the previous version better. Why do we need to summarize trademark licensing, fair use, and licenses here (of which the GPL is but one among many used for Linux components)? It is sufficient to state that they have had no legal impact on the naming question. Why spend time on a tangent that everyone agrees is irrelevant, at least currently? —Steven G. Johnson 21:41, 6 August 2006 (UTC)
I personally don't agree. There's a clear difference if for example I have no claim to your propriety or if I made no attempts to reclaim it. I would also appreciate if you'd not reverse my work and wait for a third opinion. Again, my opinion is that "no attempts were made yet" is misleading. -- AdrianTM 21:45, 6 August 2006 (UTC)
(Yes, there is a clear difference between someone trying to persuade you to do something, and someone trying to legally force you to do something. The article clearly states that this debate is of the former type and always has been. What is the problem?) —Steven G. Johnson 21:56, 6 August 2006 (UTC)

Can you give an explanation of why it is misleading to say that "No attempts have been made to legally enforce a particular name?" I clearly explained how it would have been possible for several of the parties to attempt to legally affect the naming debate had they chosen to do so, and how they could choose to do so in the future. Your continued assertion that it is "impossible" is nonsensical to me, unless you know a different meaning of "impossible" than the one in my dictionary. (If someone can do something, but chooses not to, it's not "impossible".) —Steven G. Johnson 21:51, 6 August 2006 (UTC)

Let me explain it for you. The GNU code is free, it was released under GPLv2 and now GPLv3 license. Under those license it is impossible to sue anyone that they didn't used GNU or whatever name for their products that were based on GPLed code. Now, if FSF decides to restrict the code and make a GPLv4 that demands specific naming conventions Linux will still be free to use the GNU tools that were released under GPLv2 and GPLv3, so you see it's impossible to restrict freedoms, thank God that RMS made it sure for that. -- AdrianTM 21:59, 6 August 2006 (UTC)
Sigh. I understand that, and that's not what I outlined. Suppose the GNU project (or developers of any other major Linux component) decided tomorrow (or had decided ten years ago) to change the license on all their stuff to require a "GNU/Linux" name. Yes, this is extremely unlikely and goes against the stated intentions of Stallman and the FSF. Yes, you could use the old versions, and yes, there would certainly be forking. Yes, it would be a huge mess and probably a fiasco—everyone would have to choose whether to use the official maintained version and "GNU/Linux" or some forked version by new maintainers or some replacement. Yes, it might not work (in fact, it would probably create a backlash). But is it impossible to attempt? Certainly not. (It wouldn't even be unprecedented...e.g. the old BSD license had a similar advertising clause. Suppose ten years ago GNU had announced that the advertising clause was okay and that they wanted one too.) Also, you are ignoring the trademark issue, by which Linus could easily try to legally affect the naming if he chose, especially if he was suddenly brainwashed into agreeing with the FSF position that "Linux" only refers to the kernel. As I keep explaining, you seem to be equating "no one has attempted to impose any legal restrictions" with "no one could have attempted to impose any legal restrictions." —Steven G. Johnson 22:23, 6 August 2006 (UTC)
I think we are talking about different things, I'm not sure what are you talking about, I'm talking about the legal status of the issue (current legal status, not about different speculations on the future status). Currently, GPL doesn't require GPLed derived work to be named in any way -- that's the fact, that's what I want reflected in the page. -- AdrianTM 22:35, 6 August 2006 (UTC)
I have never disputed this. But, as I explain below, this is but one corner of a more general statement—that there have been no legal attempts of any kind, not just in the GNU licenses. Second, while I have no dispute that neither the current GPL, nor any license used for Linux software, currently affects the naming debate, I object to any phrasing that implies that legal measures are impossible. I also object to any phrasing that states that there are no restrictions whatsoever on the naming, which is clearly untrue because of the "Linux" trademark. —Steven G. Johnson 22:42, 6 August 2006 (UTC)

By the way, regarding your most recent edit: if you're going to say one thing in the article, it should be the most general true statement: that there have been no legal attempts to influence the naming debate in either direction. The lack of the requirements in the current GPL (which I do not dispute, and have never disputed) is just one specific example of this, and is therefore secondary. (You keep making grammatical errors in your entries. If English isn't your first language, that might help to explain some of the misunderstandings we are having here.) —Steven G. Johnson 22:42, 6 August 2006 (UTC)

PS. Your recent edit comments went further and suggested that the current phrasing is "incorrect". You still haven't explained this. —Steven G. Johnson 22:46, 6 August 2006 (UTC)
From the first sentence of this discussion I said that that paragraph is misleading. Now, if something is misleading it's pretty irrelevant if it's "correct" on the surface. From how you put it in that sentence one would understand that there might be some legal requirements to call the OS "GNU/Linux" just that nobody chose to challenge it in court (yet), or the issue is that at least from the GNU side of things there is no legal requirement to name it one way or another. The trademark issue is a completely separate discussion and introducing it in this discussion unnecessarily complicates things (BTW, Linux trademark has already been challanged in court, your paragraph makes this issue a little bit murky).
Frankly I think that what is in the article right now is the worst variant... it's long and confusing and what's the main point in this discussion is only treated as an unclear example. The point was short and simple and presented the issue that "GNU/Linux" vs "Linux" issue is a "moral" issue (or however you want to call it) not a legal one since GPL doesn't have requirements for naming derived products. From this simple idea look what that paragraph has become....
I also don't understand why you don't have patience for a 3rd opinion on the issue. Anyway with this I end what I had to say about the issue. -- AdrianTM 03:15, 7 August 2006 (UTC)

The current bickering, and how it makes the article read

I'm opposed to the continual rewrites of the introductory paragraphs which go overboard in addressing subtle nuances to the detriment of the article's readability. Could people please keep in mind that the purpose of the intro is to lead the reader into the greater article, not to present a full summary of points.

For what it's worth, I broadly agree with AdrianTM's approach to the subject, both factually and stylistically. I'll wait until the current spat of total rewrites calms down before trying to sort the intro again though. Chris Cunningham 09:16, 7 August 2006 (UTC)

Hi Chris. My objections are not to AdrianTM but to Steven. also, as currently written, the second paragraph is the intro itself. the third and fourth paragraphs could be moved into another section; that might address your concerns. but i think those paragraphs need to be present, somewhere, as they summarize some important info that otherwise would be difficult to extract from the "arguments" sections, which present mostly quotes. Compare to the way a newspaper article works; it presents successively more detailed views, rather than jumping directly from an intro to the most detailed "he said, she said" section. (For that matter, I think the quotes need to be drastically trimmed; as it is the whole article is unencyclopedic.) Benwing 10:01, 7 August 2006 (UTC)

Yeah, I agree with all of that. The quotes section includes all sorts of not-exactly-credible sources like Mangelo as well. I just wanted to point out with each successive counter-edit the intro seems to be increasing in girth withough making any obvious impact to its accuracy. Chris Cunningham 10:50, 7 August 2006 (UTC)

I agree that the introduction should be a short summary of the basic facts and of the main arguments supported by citations/quotations later in the article. Please, however, practice what you preach, and don't insert new arguments in the intro.

The best way to keep bickering out of the intro is to keep it as short as possible. I've reduced the intro to two sentences summarizing the quoted arguments on each side, along with a short description of the composition of a typical Linux system in terms of the kernel, GNU, and other software sources. I think this should be sufficient.

(I would argue that the test for sources in an article like this is not credibility per se, it is notability and verifiability. If Bill Gates were to come out and say something completely ludicrous on the subject, it would still be worth quoting because he is notable. I don't have strong feelings on the MozillaQuest magazine quote myself, but note that this was already discussed in past Talk.)

—Steven G. Johnson 21:23, 18 August 2006 (UTC)

Quotes

The formatting change away from the over-the-top Template:cquote is much improved. This doesn't fix the problem that there are a lot of quotes for an encyclopedia article here, but at least they're not quite so distracting. --71.161.223.175 04:20, 22 August 2006 (UTC)

Because this article is mostly about a question of opinion and taste, relying heavily on quotes from notable sources—much more heavily than a typical article—is really the only way to retain any semblance of NPOV. Otherwise, it's an invitation for every editor to inject their own favorite arguments, and the article turns into a Slashdot thread. —Steven G. Johnson 15:45, 22 August 2006 (UTC)

Unfortunate

It's unfortunate that the hard working Linux crew don't have the time or inclination for nonsense like this whilst the perennial blowhard RMS fascists have infiltrated Wikipedia and sabotaged all efforts to bring about some common sense and goal oriented work in the FOSS community. All for the insatiable ego of one total jackass.—The preceding unsigned comment was added by 62.1.57.90 (talk) .

This is all because some people can't get some very simple facts. If you use free code you are free to call your resulting product whatever you like, it's not like the original author can came and claim "hey you use my code, you should give me credit", sorry man, but it's not your code anymore, it's free and I get to call my product, there's no requirement in GPL to call derived products in one way or another, if you want credit then you should use a different license that demands credit (although credit is usually shown by giving credit to the person(s) who wrote the initial code, not by keeping the name of the product, which, by the way has been modified in the meantime)... that's exactly Linus' point when he said that FSF get to name their OS however they want if they take Linux kernel and use it in GNU OS, they can even call it only "GNU" if they choose so, that's exactly what Linus did, he took free parts from GNU, put his Linux kernel and named that Linux OS other people took that and modified, added stuff and they named the resulting product: "Fedora Linux", "RedHat Linux", "SUSE Linux", "Ubuntu" and so on except Debian who chose GNU/Linux (that's their right), people in general call that "Linux", that's how most of the people call it, that's how developers call it, only RMS insists "hey, give credit to our project", now, he's right, he deserve credit, many times Linus said that he couldn't do it without GNU project, however Linus gets to name his product however he likes and people should understand that it's his product even if he used 30% free code, or 99.99%, or even if he used 100% free code, that's the nature of free code you take it and you make your own project out of it, it's really bad form from people who contributed to the initial code to come back to you and try to force you to name your product when that's not required by the license in order to get free publicity out of it -- that's really an attack on the freedom of the code, I mean they should decide if it's really free, or if it comes with additional "moral" requirements (basically free advertising) included. It's also irrelevant if some few people around think they know what people should do and try to push their POV down the throat of the majority of people, including the Linux developers and distro mainainers. -- AdrianTM 05:55, 27 September 2006 (UTC)
Don't feed the trolls. You're post contains numerous errors, many of which are answered in the FAQ. Gronky 12:23, 29 September 2006 (UTC)
Care to point my mistakes? I don't think that FAQ addresses any of the points I raised. -- AdrianTM 21:39, 30 September 2006 (UTC)
One mistake is your claim that the GNU project is "try[ing] to force you" to use a particular name - a boggling claim when you point out in the very same sentence that they have decided not to use their licence to impose this supposed "force". Another mistake is your repeated implications that FSF or the GNU project doesn't think you should be free to call the system whatever you want are incorrect. The opposite is true. FSF agree that you should be able to call the OS whatever you want - they agree that this is an important freedom. They ask (repeatedly) that you call it GNU/Linux, and they explain why (repeatedly). Asking and explaining are not "forcing" or limiting your freedom of speech/naming. Here's a transcript of an explantion of why RMS asks for credit, if you're interested, particularly the 3rd last paragraph, but the FAQ does a better job. Gronky 22:35, 1 October 2006 (UTC)
OK, It's true that they ask people (repeatedly) to call it GNU/Linux, I agree they don't "try to force" (although nagging can be sometimes perceived as a from of "forcing"), but some other people around do try to force their POV. But anyway, even "asking" people to call a project some way or another is not quite OK with me. For example I make a wonderful program that I name "Adrian" and I make it freely available for people and companies to use it, Microsoft (or RedHat, doesn't really matter who) gets that code, add their code and make a hugely popular product, then after couple of years or so I come and "ask" (repeatedly) people to use "Adrian" name because it was my idea, it was my initial code, without me it wouldn't exist, and then people who adulate me or my ideas/ideals would go around and troll forums and Wikipedia asking people to call the product "Adrian", claiming that "Adrian" is the correct name, in order to promote my ideas. Does that sound OK? I very well understand and approve the ideas that stay at the basis of GNU and FSF, however I don't think you can promote the idea of freedom by doing such things, but that's only my personal POV. Apart from that nitpicking "forcing" vs. "repeatedly asking" I don't think my argumentation is full of "numerous errors" as you said. -- AdrianTM 00:09, 2 October 2006 (UTC)