Talk:GNU/Linux naming controversy/Archive 5

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 4 Archive 5 Archive 6 →

Contents

Proposal for better disambiguation (see Talk:Linux (disambiguation))

I think GNU/Linux should be the main operating system page, Linux (kernel) should be about the kernel, GNU/Linux naming controversy should remain as is, and in fact Linux should be a disambiguation page that points to all three, saying this: Linux either refers to GNU/Linux the operating system, also known as just Linux, or Linux the kernel. For more information on this disambiguation, you can read about the GNU/Linux naming controversy. --Chris Pickett 04:34, 6 December 2006 (UTC)

I just discovered a 5th page, Linux (disambiguation). In my opinion, this should be moved to Linux. Please discuss at Talk:Linux (disambiguation). --Chris Pickett 04:47, 6 December 2006 (UTC)
I'm sorry but there are many people that don't accept the silly GNU/Linux name including most of the kernel developers. In my opinion it's "Linux kernel" and "Linux OS" (or short: Linux) , that GNU/Linux is a political motivated name that is used by RMS fans because RMS asked people to use the name to promote his ideology -- it's not by any means a widely accepted name, for one thing you'll be hard pressed to find a instance in any serious publication. Doing what you propose is a clear POV pushing. -- AdrianTM 04:49, 6 December 2006 (UTC)
Please let's just discuss this at Talk:Linux (disambiguation) where you can read my full proposal. I responded to you in another message that I'm not trying to push a particular POV. --Chris Pickett 05:00, 6 December 2006 (UTC)

Linux 0.12 under GNU GPL?

No, Linux 0.12 was certainly not released under the GNU General Public License. Linus Torvalds, the principal developer of Linux, just suggested switching to it. -- mms 01:49, 15 January 2007 (UTC)

Linus Torvalds in the release notes of Linux 0.12: The Linux copyright will change: I've had a couple of requests to make it compatible with the GNU copyleft, removing the "you may not distribute it for money" condition. I agree. I propose that the copyright be changed so that it confirms to GNU - pending approval of the persons who have helped write code. I assume this is going to be no problem for anybody: If you have grievances ("I wrote that code assuming the copyright would stay the same") mail me. Otherwise The GNU copyleft takes effect as of the first of February.

Reverting of the table form

Ok, I spent at least an hour fixing that up, and I think it's way easier to read in table form. I was inspired by A Defense of Abortion which I consider beautifully formatted. I think article as you've reverted it is very difficult to read. Please consider letting the table form stand for a while so that others can see what it looks like. [1] Chris Pickett 04:51, 19 January 2007 (UTC)

Recent major changes

User:Chris Pickett recently made several major changes to the article that I think are unwarranted, and in any case should have been discussed here first considering that this article has been fairly stable (and featured) for a long long time now.

  • First, it's not appropriate to format almost the entire article in the form of a table. Wikipedia (and standard encyclopedia) style, not to mention standard writing style in general with the exception of some legal and technical documents, is to use prose for the majority of discussion.
    • Except that it wasn't prose, it was argument, quote, argument, quote, argument, quote. If you can provide a link to WP official policy that would be great. Chris Pickett 05:02, 19 January 2007 (UTC)
  • Second, this article needs a 1-paragraph summary of the history, which is what it had and has. It can link to the Linux and GNU articles for more information, but (1) those articles are not focused on the connection between GNU, the Linux kernel, and complete systems (whatever "complete" means), and (2) the history is rather central to many of the arguments and needs to be summarized for the reader to comprehend what is going on.
    • I merged that history into the Linux history, since it was 80% repetition. There are several people who agree that moving this article to GNU/Linux as an article solely about the name GNU/Linux is a good idea. From both sides of the fence. See Talk:Linux. Perhaps a much shorter summary of that history will do. Chris Pickett 05:02, 19 January 2007 (UTC)
  • Third, as a minor point, the "pronunciation" section is a rather minor side issue and doesn't belong up front
    • Okay, I was just following style I've seen in other articles (in fact, usually it's in the lead sentence). Chris Pickett 05:02, 19 January 2007 (UTC)
    • I've been working hard to sort out the GNU/Linux mess and just because an article has been in a totally unreadable format for a long time doesn't mean that's the best format. Please assume good faith here, I'm not an idiot.

—Steven G. Johnson 04:53, 19 January 2007 (UTC)

One thing that you need to realize is that this article has to stay unusually close to quotations, because the topic is so controversial among editors (witness the endless battles on Talk:Linux). Otherwise, it degenerates very quickly into a Slashdot-like thread with every editor inserting their pet arguments ("some people say..."), and indeed this is what happened with early versions of the article. Only by insisting that every argument be directly tied to a prominent commentator did the article become reasonably stable (and, according to the featured-article debate, at which point the article had pretty much the same structure and no one objected, "the NPOV tone is exemplary"). —Steven G. Johnson 03:33, 12 February 2007 (UTC)
There's no question that first paragraph of the history section is a summary of the longer history sections in Linux and GNU. However, this is standard practice in Wikipedia, as well as being common sense—we have a brief one-paragraph summary of information directly relevant to the present article, and link to the longer article for a more full description. I'm not sure why you think this is a problem. —Steven G. Johnson 03:33, 12 February 2007 (UTC)
I do realize that it must stick to quotations, and I've seen the Talk:Linux arguing. Also, you're right, the history section is actually necessary to provide context. As I said above, I think what could help this page a lot is to move it to GNU/Linux; that would put an end to revert wars where wikilinks alternate between GNU/Linux and Linux, it would make GNU/Linux a featured article, and generally I think it would help to stamp out flame wars. There's that, plus the actual identification of arguments which I mention below. Chris Pickett 22:15, 16 February 2007 (UTC)

Free Software Foundation?

I can't find any evidence on the FSF site that they officially promote the term GNU/Linux. Is it the GNU project that is meant here? Neither the GNU project nor Richard Stallman are the FSF. 80.233.255.7 18:29, 19 January 2007 (UTC)

One of the FAQ's is copyright the FSF. Chris Pickett 06:16, 21 January 2007 (UTC)

New Subections

Barring a table (bad idea according to Stevenj), I think breaking each section into subsections labelled with the argument being discussed would be very helpful and allow the reader to get a summary of all the arguments quickly. For example, you could have === The name Linux is ambiguous === as a subsection of Arguments for GNU/Linux, and move the corresponding discussion and quotes into that section. Presently the reader needs to read all the text and decipher the actual arguments. Chris Pickett 06:16, 21 January 2007 (UTC)

A subsection marker causes more harm than good if the subsections are too short, as they will be in this case (with one or two quotes and one or two sentences per subsection, I'm guessing). For readers wanting a summary, I would argue that the best thing is to have a good introductory paragraph to the sections summarizing the arguments. —Steven G. Johnson
Hi Steven. Thanks for your responses. I guess my biggest problem with the article is this: I often can't figure out what specific argument each particular quote actually makes. It is very he-said she-said and I'm kind of surprised that it made it to FA status. I think it should be possible to determine the precise argument(s) pertaining to each specific quote without introducing bias. Chris Pickett 21:58, 16 February 2007 (UTC)
To follow-up, I think \paragraph{<argument>} style formatting would work, e.g.:
GNU/Linux is not about RMS' personal ego. In response to suggestions that Stallman's renaming efforts stem from egotism or personal pique, RMS has responded that his interest is not in giving credit to himself, but to the GNU Project:

Some people think that it's because I want my ego to be fed. Of course, I'm not asking you to call it "Stallmanix".

Acceptable? I think a summary is also appropriate; this just helps identify the exact arguments at the particular location in the text. It also makes it clear where there is possibly extraneous text: for example, it seems to me that ", but to the GNU Project" above is described elsewhere and not supported by the quote. This is an example of where I get confused, trying to match up the article text with the precise quote text. (I also think counter-argument quotes like this should have matching original argument quotes, as well as follow-ups. To me, the inclusion of this quote alone is somewhat biased: there is quite possibly somebody notable who said that even though he doesn't want it to be Stallmanix, GNU still feeds RMS' ego since GNU was his creation.) Chris Pickett 00:16, 17 February 2007 (UTC)
The purpose of a header, like a summary, is to let people quickly get the gist of a longer passage. In your example above, the only purpose of your bolded sentence seems to be to summarize the following sentence. Having one sentence summarize the next one is a little superfluous, don't you think?
It summarizes the sentence and the quote in seven words. Yes, it's a short example, but I think having a bolded summary like that for each individual argument in the article would sure make it easier to grasp what they all are in say, less than one minute. Don't you think?
Below, you are arguing "to avoid the redundancy where possible", and here you are arguing to summarize a longer sentence with a shorter one immediately preceding it. My head is spinning. =) Seriously, summaries at the beginning of a long section are good, but who uses a summary for an individual sentence? If you want a summary, why not propose a one-paragraph summary for the beginning of the section? —Steven G. Johnson 18:32, 20 February 2007 (UTC)
I'm not arguing for a single summary sentence for each sentence, and the example I chose was quite possibly the worst one possible. (I chose it because it was short---look where that got me.) Anyway, let's do something constructive. Here are the main arguments from the GNU/Linux section:
  • GNU is a larger technical contribution than Linux. (First sentence of quote 1 and quote 11.)
  • GNU project set out to make a complete OS. (Sentences 5 and 6 of quote 1.)
  • GNU idealism played an important role. (Quotes 2 and 3.)
  • GNU and Linux are necessary for complete Unix functionality. (Quotes 4 and 4.5---the one in the body text.)
  • Linux by itself is confusing. (Quotes 5, 6, 7, 8.)
  • GNU/Linux is not about RMS' personal ego. (Quote 9.)
  • GNU/Linux gives fair credit to the GNU project. (Quote 10.)
It took me 20 minutes to put this list together, and I've read the article a few times. It is even more clear to me now how these few short summary sentences in bold will serve readers; maybe they should even be === headings, but that's minor. I also might have missed something, I'd appreciate if you can comment. I realize now that there isn't actually a part that says it's explicitly about giving GNU credit, and that should be the first quote, even though it's bash-the-reader-over-the-head-obvious. ("I'm asking you to call it GNU [sic], because I want the GNU Project to get credit.") I also think that quote 1 should be split into two, so that the first argument in my list can be grouped with quote 11 and the second argument can be grouped with the last part of quote 1. Finally, I think quote 10 and its argument should be moved up to the top with the new "I want the GNU project to get credit." quote.
Here are the main arguments from the Linux section:
  • Linux is the most widespread name. (Quote 1.)
  • Linux is an older name. (There is no corresponding "the GNU project is actually older" argument, which would only be fair.)
  • Linux is shorter and easier to pronounce. (Pronunciation is a forward reference here...)
  • GNU Linux [sic] is only justified if GNU makes a Linux distribution. (First part of quote 2.)
  • GNU/Linux is about RMS' personal ego. (Quote 3.)
  • Linus thinks Linux is the right name. (Last part of quote 2, last part of quote 4.)
  • de facto argument: nomenclature is irrelevant. (First part of quote 4, quote 5, last sentence of quote 6.)
  • Claiming credit makes GNU look bad. (Quote 5.5, in the body text---should be a full quote.)
  • GNU is not a more important contribution. (First sentence of quote 6.)
  • Linux is a standard example of free software. (Pretty much the same as argument 1, should be grouped.)
This also took me 20 minutes. It's just not totally obvious. Did I miss anything? Here, I agree that summary sentences / subsection headers would be more difficult to apply as the article stands, but I also think that draws attention to the fact that some arguments need more weight given to them. (Ordered by priority, arguments 1, 3, 2.)
I'm just trying to make sense of this all. It's finely crafted and well-cited text, but I don't think it's particularly modular, and organizing it all by specific argument would help that. (I find the GNU/Linux section is better in this respect.) Chris Pickett 21:05, 20 February 2007 (UTC)
Regarding "but to the GNU project" being "not supported by the quote", this is obviously what RMS meant in context (where he explicitly says, "I'm asking you to call it GNU, because I want the GNU Project to get credit."). Whenever we quote a short passage from a longer source it is our obligation to put the passage in context.
I still think it's better to avoid the redundancy where possible, since giving credit to GNU is the primary argument for GNU/Linux and already made elsewhere.
Above, you were saying it's "not supported" by the quote, which means that this point was not obvious to you from the quote...this is precisely why we need to provide that context here (which is explicitly in the original source as I mentioned) in order to clearly convey what the speaker meant: he was making a distinction between credit to himself (Stallman) and to the project (GNU) which he started, a distinction that not everyone sees or agrees with. —Steven G. Johnson
Yes, see above, that argument is so important it needs its own quote. Chris Pickett 21:05, 20 February 2007 (UTC)
If you can find a prominent commentator who has responded to RMS's "Stallmanix" quip (which itself is in response to remarks like the Linux Journal editorial we already quote...), please let us know so that we can include a reference to this response, if it seems appropriate. —Steven G. Johnson 21:28, 19 February 2007 (UTC)
Well, basically I think there are some quotes that aren't really worth including, this one about Stallmanix being one of them: Stallman arguing that it's not about his ego just because it isn't called Stallmanix isn't the most sound argument, but I suppose that's just my opinion. On the other side, I'd argue that Linus saying it would be appropriate for it to be called GNU Linux if the GNU project actually produced its own Linux distribution is quite clearly disingenuous, again my opinion. On an off-topic note, I'm as much an editor as you, and "please let us know so that we can include a reference" isn't particularly welcoming. Chris Pickett 09:37, 20 February 2007 (UTC)
The reason I included the Stallmanix quip is because it was the only place I could find where Stallman directly responds to the accusation (which we include) that "GNU/Linux" is mainly an ego thing. Can you find a better quote? I'm not sure it's our place to decide whether his rejoinder is convincing or not. —Steven G. Johnson 18:32, 20 February 2007 (UTC)
I looked through the first page of hits on Stallmanix but nothing much. Chris Pickett 21:05, 20 February 2007 (UTC)

New "Arguments for neither" section

There seem to be some quotes that suggest the whole naming controversy is a waste of time, and I don't think they should be lumped into the Arguments for Linux section. The one with Linus saying personally I'll very much continue to call it Linux is such a quote (unless you count it as an Argument for Linux by Appeal to Penguin Authority, which, um, isn't exactly good), and there's another one about the community wasting too much time on nomenclature. Chris Pickett 06:16, 21 January 2007 (UTC)

An argument that the controversy is a waste of time is an argument for maintaining the status quo, and thus is de facto an argument for "Linux". (Indeed, it's quite a strong argument from a certain point of view, since most language is an somewhat arbitrary consequence of history and convention, and English contains any number of usages that are rather illogical but are too entrenched to change.) —Steven G. Johnson 03:08, 12 February 2007 (UTC)
That's a good point. Chris Pickett 21:59, 16 February 2007 (UTC)

Quote attribution

I think each quote should end with (full name, organization, year), before the reference. Some of them already do, but not all. I had this in my latest revision, in fact, you can dig out the info from there. (Don't revert, it has the tables which are no good.) Chris Pickett 06:16, 21 January 2007 (UTC)

This is essentially Harvard referencing, and used to be the convention in the article, but was switched when the article was moved to the new <ref> numbered-endnote notation. However, I totally agree that the information should be added back into the text of the article at least in any place where it isn't clear who is saying what. —Steven G. Johnson 03:13, 12 February 2007 (UTC)
Maybe a combination of two systems would be good, per Finnish Civil War. Each reference actually contains several arguments, and a seperate footnotes section that maps quotes in the article to full references would be useful. Chris Pickett 22:05, 16 February 2007 (UTC)

Maybe...

Maybe GiNUx ? AtomicZero 04:31, 28 January 2007 (UTC)


or GNUX, (pronounced nuks) "Building Viewing Using computers in GNU & Linuxiting ways" i've made graphics for it too

gnux overlay
gnux overlay
gnux 2
gnux 2
gnux caption
gnux caption

these images are designed to appeal to a new user, informing them and becoming a recognisable conjunction, easily reffered to in one syllable without destroying the derivatively decipherable terms recognisable by experienced users. oh, and you can have my images under gfdl ok?

Michael hatton 09:03, 2 March 2007 (UTC)

LiGNUx? HoCkEy PUCK 21:19, 2 May 2007 (UTC)

BSD advertising clause

Someone inserted a long quotation regarding RMS's well-known opposition to the BSD advertising clause into the "Arguments for Linux" section, implying that he is somehow contradicting himself. I removed it because:

  • First, it is unsourced. While no one disputes that RMS opposes the BSD advertising clause, there was no citation of any prominent commentator using this as an argument for "Linux". This article is not for editors to insert their own arguments, whether good or bad; otherwise the article degenerates quickly into a Slashdot thread. We must stick closely to arguments that can be sourced to prominent commentators.
  • Second, the argument is obviously fallacious. RMS argued against BSD's legal requirement of credit in advertising materials (on practical grounds), not against giving credit in general; he has never advocated that people be legally required to use the "GNU/Linux" name.

—Steven G. Johnson 02:59, 12 February 2007 (UTC)

I've found several people arguing that this is hypocritical. I don't know if you think Terry Hancock is notable enough for Wikipedia or not, but there may be other more "notable" sources that have the same opinion. [2] Chris Pickett 03:11, 1 March 2007 (UTC)
Certainly Theo de Raadt is notable (and relevant to this particular argument) [3] Chris Pickett 03:31, 1 March 2007 (UTC)
And if it's unfair to quote this BSD-related passage from RMS, I don't think quoting Linus in the "Arguments for GNU/Linux" section is appropriate either: it's quite clear he doesn't support the name, and he wasn't making an argument for GNU/Linux. Or maybe it's better to include both as examples of perceived hypocrisy. Chris Pickett 04:01, 1 March 2007 (UTC)
de Raadt is certainly notable, but the link you cite doesn't mention the advertising clause per se—it is a vague accusation (without citing any evidence) that the FSF "nearly goes out of their way to avoid citing others". Terry Hancock is not nearly so famous, nor is the source you linked (a random response to a blog post).
It was in a thread about GNU/Linux and I figured it was fairly obvious from the BSD context, but I also agree it isn't the best source. Chris Pickett 08:43, 1 March 2007 (UTC)
I didn't say that quoting RMS per se was inappropriate—quoting anyone is fine to establish facts or if the quotes directly reflect on prominent arguments. However, all I was saying was that we shouldn't quote someone to make our own arguments, but rather to illustrate arguments that appear in prominent sources. The Torvalds quotes reflect directly on the FSF's explicit argument that a kernel is only a small part of a complete system, and hence they are clearly relevant. In a similar way, if RMS ever said "I admit, GNU/Linux is awfully inconvenient to say" this would reflect directly one of the common arguments for "Linux" and would therefore be relevant to quote.
Well, I don't really think it was an admission that Linux was possibly not a good name on Linus' part, and everybody's going to agree on facts. Anyway, whatever, it's fine. Chris Pickett 08:43, 1 March 2007 (UTC)
If someone prominent indeed made this argument about the BSD clause, or someone writing in a citable source (e.g. a magazine or other publication), we should definitely mention it. However, we also have an obligation to mention that it implies an obvious distortion of Stallman's position: RMS has never argued against giving credit in general (and indeed, every license requires that the copyright holders' names be preserved), only against the legal requirement in advertising materials, nor has he ever argued that people be legally required to use the name "GNU/Linux". We have an obligation to report arguments made by prominent commentators, but neither should unambiguously misleading/nonsensical arguments be reported uncritically.
—Steven G. Johnson 07:40, 1 March 2007 (UTC)
Yes, I agree it is obviously not what RMS is arguing for, but I don't think it's entirely unreasonable to accuse him of wanting to have his cake and eat it too by going around insisting on GNU/Linux everywhere (along with many devoted followers) while also insisting that this not be legally enforceable. It's your opinion that it's nonsensical; it's my opinion that it's not. Chris Pickett 08:43, 1 March 2007 (UTC)
The cornerstone of a free society is that we believe others have the legal right to express opinions that we disagree with. ("I disagree with what you say, but I will defend to the death your right to say it.") Saying that RMS is somehow hypocritical to simultaneously argue for "GNU/Linux" and yet not try to legally enforce this name is rejecting that cornerstone. In fact, it is implying that everyone who believes in freedom of speech is a hypocrite—how dare we disagree with someone's opinion, and yet not try to outlaw their expressing it! Surely, that's not really what you think?
It's not the "not trying to enforce it"---it's insisting that others not be able to enforce something, yet going around insisting on almost that exact same thing (i.e. credit) without a legal document, and essentially obtaining it through brow-beating. And I don't necessarily fully agree either---it just doesn't strike me as non-sensical. Like I said, the non-sensicality is your opinion. Chris Pickett 09:33, 1 March 2007 (UTC)
Even if that is your opinion, can we find a prominent commentator making that argument?
Not yet, I only tried for 10 minutes. Anyway, I give up for a while. Really time to See You Later now :) Chris Pickett 09:33, 1 March 2007 (UTC)
—Steven G. Johnson 09:05, 1 March 2007 (UTC)

New arguments from Alan Cox

I also found some arguments against GNU/Linux from Alan Cox. [4] Chris Pickett 03:11, 1 March 2007 (UTC)

His main point seems to be that, "Linux [the kernel] is not and never has been an FSF project. I would say the majority of the kernel developers don't buy the FSF political agenda." I'm not sure that this is worth quoting because there is no controversy on this point—the FSF and Stallman have explicitly agreed that the Linux kernel itself is not a part of GNU. And our article already makes this point in several places.
If we want to quote Alan Cox, I think there are other posts where he talks about the GNU/Linux naming issue in more detail (whereas this post is mainly about kernel licensing). It is well-known that Cox advocates the "Linux" name for Linux-based systems. For example, see this post (to be fair, RMS responded directly in another post). The only question in my mind is whether this exchange adds any substantially new arguments or new information that we do not already summarize. —Steven G. Johnson 08:08, 1 March 2007 (UTC)
In general I think threads where Stallman and other notable people have commented directly are worth linking to, if not necessarily quoting from.
I apologize for not explaining myself. As far as I see it, Cox's main argument is that there is no such entity even called GNU/Linux, and hence the whole Linux kernel GPLv3 uproar really has nothing to do with the FSF, and Linux isn't splitting from the FSF or anything like that at all. This is an argument that GNU/Linux causes confusion by associating GNU with the Linux kernel. (c.f. arguments about SCO confusing the OS and kernel). There is a second argument that he mentions parenthetically: it is a trademark issue. I don't see either of these arguments in the article text. Chris Pickett 08:28, 1 March 2007 (UTC)
You're right, it's an interesting counterpoint: he's saying that using the term "GNU/Linux" confuses the FSF with issues solely to do with the kernel. (And indeed, the FSF has been quick to make a similar point, that the kernel licensing has nothing to do with it or the rest of the GNU software.)
Regarding his statement that there is no such entity as "GNU/Linux", that seems to be more a statement of his conviction/conclusion than an argument per se...he doesn't really explain or justify it. (Obviously combinations of the GNU components with the Linux kernel exist; his opinion is clearly that this should not be considered an "entity" or named "GNU/Linux", but stating so isn't an argument. Nor does he intend it as one, probably...his main point in the posting seems to be to make clear that the FSF is not involved with the kernel.) —Steven G. Johnson 09:14, 1 March 2007 (UTC)

Interesting article by Hans Reiser

I thought this was sarcastic at first but apparently he's serious! [5] Chris Pickett 03:23, 1 March 2007 (UTC)

This is a well-known part of Reiser's philosophy, I think. This is a guy who named a filesystem after himself, and if I remember correctly advocated naming components after their major developers. He chips in on the FSF's behalf, "When Richard Stallman isn't even mentioned on the box as an author because doing so does not further the mindshare capture effort of the distro, well, this is just wrong." However, this is the extent of his comments on the issues, and merely saying that something is "just wrong" doesn't add much more than a "me too". I'm not sure it's worth quoting. —Steven G. Johnson 07:47, 1 March 2007 (UTC)
I just thought it was interesting because on one hand you have RMS saying it's not about ego, and on the other you have Hans Reiser calling it GNU/Linux and saying, "Why should it be about anything but ego?" I don't say I agree personally, but I also find it's a much nicer defense to a criticism of egotism than, "I'm not egotistic (gr?), this is why." Chris Pickett 08:14, 1 March 2007 (UTC)