Talk:GNU/Linux naming controversy/Archive 7

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 1

←

Archive 5

Debian naming issue

I've added the Naming conflict between Debian and Mozilla link to "See also" because it's one of a very small number (3?) of articles about naming issues in the free software community. It's reasonable to assume that people interested in one naming issue might be interested in other naming issues. --Gronky 08:08, 23 October 2007 (UTC)

It's off-topic (perhaps you may consider developing an appropriate topic which covers both conflicts) Tedickey 10:22, 23 October 2007 (UTC)

Why is one free software naming conflict off-topic in an article about another free software naming conflict? --Gronky 10:41, 23 October 2007 (UTC)

There are a lot of them - and the place to discuss roughly-similar issues is in a topic devoted to this rather than wandering around collecting things that might be related in case someone wants to organize the material Tedickey 11:17, 23 October 2007 (UTC)

Are there a lot? If there are, I must have missed them. Can you give me the links? I'm only aware of Naming conflict between Debian and Mozilla and alternative terms for free software, so to me it looks like a topic with only three articles. --Gronky 11:20, 23 October 2007 (UTC)

In more than one direction: Mozilla has been the subject of naming conflicts for more than one operating system for instancs. There are several conflicts regarding ownership of names (and relabeling other people's work to take credit for it). You won't find many neutral discussions of those however (wikipedia tends to reflect the majority opinion rather than citing facts). Tedickey 11:25, 23 October 2007 (UTC)

But there are no Wikipedia articles about any of these conflicts, are there? I recommended adding a link, you disagreed because "There are a lot of them", but it seems there are only three. There are surely more that have happened that don't have Wikipedia articles, but no one's suggesting adding those to the "See also" section. --Gronky 13:25, 23 October 2007 (UTC)

I'm thinking that maybe a category about naming would me more appropriate than links to different articles. -- AdrianTM 13:01, 23 October 2007 (UTC)

But there are only three articles on this topic. At category would be overkill. Adding two links to the "See also" section of each article is not a big deal. --Gronky 13:25, 23 October 2007 (UTC)

It's not a big deal, but it's only very vaguely related, it's a naming issue, not even a "controversy", and in this case it's clear who controls the names, and Firefox or Mozilla don't have any special connections with either GNU or Linux. -- AdrianTM 14:06, 23 October 2007 (UTC)

OT: FWIW, I don't think this article should have "controversy" in the title anyway. I previously suggested changing that word to "debate", but someone disagreed with changing it because they wanted to keep the title that the article had when it achieved featured status (which it still had at the time of my previous proposal). --Gronky 21:39, 25 October 2007 (UTC)

To me "debate" sounds like that there are two parties that sit at a table and discuss... to me this sounds way too civilized for what it actually happens -- in this case one party claims something and the other one is either dismissive about that or plainly ignores those claims... and there's yet another party (the large number of people) who couldn't care less. -- AdrianTM 22:01, 25 October 2007 (UTC)

"Mepis Linux"

I couldn't find any evidence on the distro's site that "Mepis Linux" is the official name of the distribution. Most of the time it's referred to as "MEPIS". 80.233.255.7 10:05, 15 November 2007 (UTC)

Perhaps not an official name, but the most prominent thing on the website is "Why SimplyMEPIS Linux?". I don't know exactly what the threshold is for what a distro's "official name" is, but I would say that's fairly close. cf. gNewSense, which, given its origins, would definitely identify itself with "GNU/Linux", does not have "GNU/Linux" in its "official name (gNewSense)" either: it instead says it is a "GNU/Linux distribution" on its website.

Anyways. I would add a reference, but I am somewhat reluctant to link to the front page for such detail, since that's almost bound to change a few months from now. novakyu (talk) 08:36, 29 January 2008 (UTC)

Windows- originator of general public's concept of Operating System?

I'm sure many contributors to this article would wish that this article makes no reference to Microsoft or Windows. I think in the general public's mind, however, Microsoft set a precedent in 1990 when it marketed and sold Windows 3 and 3.1 as an Operating System. Really it was a sort of windowing system on top of MS-DOS, and it was fairly trivial to install it on top of DR-DOS and other versions of DOS. "In the version of MS-DOS first shipped as DOS-Windows 95, the following changes to the above DOS boot process were made: The DOS kernel image became a single file, io.sys, containing what was previously split across io.sys and msdos.sys. This allowed the name msdos.sys to be re-used. That file became a text file containing various configuration directives that the DOS kernel obeyed." from http://homepage.ntlworld.com./jonathan.deboynepollard/FGA/dos-windows-boot-process.html ^[1]

Now, many technically saavy users brought this up as a point of ridicule against Windows as another example of deception by Redmond, and it was fairly simple to see the loading of DOS when Windows95 and Win98 boot screens were turned off. The situation was made worse because Microsoft had been announcing their intention to introduce WinNT server code into their consumer PC OS before 95 was finally released.

But, They did set this fuzzy definition of operating system for the general public, and the young Computer Science student who grew up with Windows in their schools- In its early days linux became popular specifically as an alternative to Windows, and the difficulty and hazards of developing software for Windows. Macintosh of course was another alternative, but expensive hardware (which created a reputation they have never been able to shake) and lack of command line in Mac OS (for end-users anyways- despite workarounds for developers, all end-user applications needed cumbersome GUI interface) I think such factors Drove people to GNU/Linux-linux. When Microsoft equated DOS+Windows3.x as a new operating system- logically GNU+Linux was accepted as "linux."

Obviously as "original research" this isn't acceptable as Wikipedia article-grade material-- but I do think this Windows naming convention goes a long way toward explaining why the media, internet users, etc. found it natural to refer to the entire operating system with utilities as Linux. Cuvtixo (talk) 01:59, 15 December 2007 (UTC)

I disagree. Linux is more like DOS and GNU is more like Windows in your analogy. The user sees GNU but he doesn't see Linux (assuming Linux is just a kernel). --mms (talk) 14:56, 15 December 2007 (UTC)

I disagree, KDE or Gnome are like Windows, GNU is like DOS tools and Linux is like Windows kernel -- AdrianTM (talk) 17:36, 15 December 2007 (UTC)

I've never heard of an operating system called "DOS tools". Is there a Wikipedia article about it? :-/ --Gronky (talk) 14:07, 14 January 2008 (UTC)

DOS utilities, some of the DOS commands are internal, some are external (not part of the DOS kernel), for example GNU command "less" is a lot like "more" command in DOS, "ls" command in GNU has a parallel in "dir" (internal I think) DOS command, don't remember DOS very well to give you other examples, but I think the parallel is pretty clear, GNU is pretty much like DOS while KDE and GNOME are like Windows, as I said in response to mms. I said "DOS tools" because now in Windows I'm not sure what they are "more, dir, cd, del" are bits of DOS legacy... -- AdrianTM (talk) 10:18, 29 January 2008 (UTC)

Reducing quotes

The page has been rightly tagged for having too many quotes. I think a lot of the quotes could be folded into references, like I did in this edit: [1]

This avoids wasting any research done by previous wikipedians, and it makes the text less redundant. "Many think cows are big, Stallman even said 'I think cows are big', Linus replied 'Cows can indeed often be quite big', etc" These quotes could be references. --Gronky (talk) 14:00, 14 January 2008 (UTC)

Yep. More of the same please. Chris Cunningham (talk) 14:07, 14 January 2008 (UTC)

Per the request and the tag, I removed all block quotations from the pro-"GNU/Linux" argument. I would be glad to do the same for "Linux" side, but given that I, as a member of FSF, am much more fond of "GNU/Linux", I am bound to remove too much and get into the whole POV issue. Could someone supporting the use of term "Linux" for the free operating system clean up that side, please? novakyu (talk) 08:19, 29 January 2008 (UTC)

"As a member of FSF" you should be careful how you edit this article, given the conflict of interest. -- AdrianTM (talk) 10:31, 29 January 2008 (UTC)

Er, that's exactly what I did here. I believe since this is "tell both sides of story" kind of article, I am free to edit the pro-"GNU/Linux" argument, as long as I don't add anything unsupported (and if anything I actually removed stuff), and I kept my hands off from the pro-Linux side (except for removing a few poorly sourced sentences and paragraphs that anyone is free to add back in, as soon as a reputable source (that is not misleading, like the one to Free Software Magazine) is found for such claims.

I am, at least, disclosing all my sources of "conflicts of interest" (it's as much as "conflict of interest" for a registered Democrat to be editing Wikipedia article about Democratic party—i.e. not much. I am not a leader or a significant member, I am just a recent recruit, and I have no financial gains to be had from promoting FSF agenda, unlike many companies that do push "open source" over "free software"). Who else here has secret agenda that they have not disclosed? Anyone? novakyu (talk) 21:38, 29 January 2008 (UTC)

Since this is an issue that obviously arouses strong feelings, directly quoting statements by prominent persons, rather than inserting our own paraphrasings, has been a good way to keep out of trouble with respect to random Wikipedians inserting their own personal favorite arguments on one side or the other. Thanks in a large part to sticking close to such quotations, I think, this article has remained reasonably stable for some time now, at least in the "opinions for X" sections. And frankly, I don't understand the "too many quotations" objection...we're only talking about a dozen quotations here. And, as a reader, it's much more informative to hear directly what Jim Gettys or Richard Stallman or Linus Torvalds etc. has to say about the issue than some random Wikipedian's imprecise and opinionated paraphrasing.

—Steven G. Johnson (talk) 21:57, 29 January 2008 (UTC)

The problem is, this doesn't magically fix all the controversial issues. You can quote whole paragraphs, or even entire articles, and still insert your own bias by quoting them out of context, etc. Sure, the possibility may be more limited than when you are free to paraphrase, but it still can be done. For example, I believe some people here think that the quotation of Torvalds in the pro-"GNU/Linux" is unwarranted (especially given that those were for an unrelated topic, and nearly a decade and a half ago). Anyways. I just want to say, having a whole lot of direct quotation is not a panacea to the controversy, and given its ineffectiveness, we should give priority to the Wikipedia policy (whatever that is—I haven't seen one quoted, or a discussion where such consensus was determined). novakyu (talk) 22:24, 29 January 2008 (UTC)

Except that direct quotations have proven very effective here. The amount of flamewars or changes in what is actually said in the "opinions for/against" section have been very limited when it comes to the quoted statements (that is, there have been lots of arguments on the Talk page, but the spillover into the opinions section of the article has been very limited...those sections haven't changed substantively for some time, which is another reason to be cautious about changes for supposedly aesthetic reasons). (In contrast, you should have seem what the article was like before we were sticking closely to quotations.) [Instead, people try to argue by proxy ("too many quotations is ugly").] As for the age of the quotations, unless you can find a source to indicate that someone has changed their opinion, that hardly seems relevant—10 years is not all that long a time (and it certainly doesn't seem that many opinions of prominent commentators on this subject have changed much in that time). —Steven G. Johnson (talk) 22:36, 29 January 2008 (UTC)

Re: removed initial sentence from pro-"Linux" argument

I just removed:

"Linux" is by far the most widespread name,^[2] and most people therefore simply adopt this usage, while references to the naming controversy appear only infrequently in mainstream sources.

for two reasons:

Google Trend is not a valid source. It amounts to original research, and furthermore, incorrect research. Google Trend collects statistics on the searched terms. That does not mean more people use the searched term. Terms like "sex" used to be searched a lot. Did that mean people used word "sex" more than any other word in their life? No.
It is not true. Here is a perhaps fairer comparison. In the Googlefight style, here's the Google search for "GNU/Linux" (at this time of writing, I get 30.6 million results). And here is the Google search for Linux -"GNU/Linux" (showing, for me, 54.6 million results). The -"GNU/Linux" is important, because if you simply search for "Linux", you are bound to pick up GNU/Linux as well (hence ensuring that Linux has always the upper hand). Also, when someone mentions GNU/Linux along with Linux, then it's quite likely he's using Linux as a mere shorthand for GNU/Linux (much like "America" vs. "The United States of America", although here the relationship is reversed, since strictly, America includes U.S.A., not the other way around), so we should discount that as well.

According to the above results, while it is clear that Linux is more often used, it's not "by far the most widespread"---it's not even twice more widespread than GNU/Linux, and since the statement proposed is not obviously true (I don't mean "obviously not true", I mean "not obviously true"), it is best not said on the Wikipedia to maintain NPOV. novakyu (talk) 09:20, 27 January 2008 (UTC)

Let me point that searching for Linux bring 824,000,000 results, it's obvious that "Linux - GNU/Linux" doesn't bring enough results (I don't know why). Simple mathematics GNU/Linux = 30 million results, if you take out 30 millions out of 824 millions you still have around 800 millions of Linux (without GNU/Linux) results. -- AdrianTM (talk) 10:23, 29 January 2008 (UTC)

Another google number is that there are 335 million hits for "gnu" and 745 for "linux". Since most user of GNU are people running the Linux kernel, and most users of the Linux kernel are running GNU, this indicates that both are in the same order of magnatude of interest to each group. "Linux" could be inflated by its being the most problematic part of the OS, where most tech discussion is needed. --Gronky (talk) 10:34, 29 January 2008 (UTC)

Let's not play with Google too much, gnu is also a name of an animal. Most importantly the debate is not between GNU and Linux, we talk about GNU/Linux vs. Linux here, that has a different order of magnitude: 30 vs. 800. -- AdrianTM (talk) 10:40, 29 January 2008 (UTC)

Yes, but how many of those 800 are about the kernel? Unfortunately, increasing the complexity of the search (per your "linux -gnu/linux" search suggestion) doesn't seem to increase the precision: Linux is also the name of a washing powder, but if you search for "linux -washing", you only get 53 million. Do we conclude that the other 700 million hits for "linux" are about the washing powder? --Gronky (talk) 11:17, 29 January 2008 (UTC)

The problem here is actually sourcing. You know you're in trouble when you're actually using a Google result as a reference. We should find a secondary source which backs up the statement that "Linux" has the weight of popularity behind it. This really shouldn't be too difficult. Chris Cunningham (talk) 11:01, 29 January 2008 (UTC)

It is actually quite difficult to find references to the Linux vs. GNU/Linux debate in the mainstream press, except in interviews with Stallman, relative to the number of discussions of Linux—that why it is so hard to find press sources for this article, because most authors don't even take the name as a question. Anyone can easily verify this. And it should be obvious to anyone with an interest in the subject that "Linux" is the most widespread term; this is already taken as a given in Wikipedia — that's why we title the article on the OS Linux instead of GNU/Linux, regardless of the technical arguments to one side or the other. Even the GNU project concedes this: the version of GNU which is widely used today is more often known as “Linux” is the second sentence of RMS's "Linux and GNU" essay [2]. At a certain point, one has to concede the obvious. —Steven G. Johnson (talk) 16:44, 29 January 2008 (UTC)

That's a quite circular argument there: "that's why we title the article on the OS Linux instead of GNU/Linux, regardless of the technical arguments to one side or the other." In other words, Wikipedia says "Linux" is more often used than "GNU/Linux" today, therefore we should state in this article that "Linux is obviously far more widespread" (I, of course, as someone with a great interest in the subject, do not think that's true, except in colloquial usage (i.e. "America" vs. "the United States of America"), but you should take that with a grain of salt). As for the article that you mention, that was written, what, more than a decade ago? Since the explicit goal of the article was to get people, and especially reputable free software publications and organizations, to say "GNU/Linux" instead of "Linux" as the full name of the free operating system, hopefully the trend has changed by now? Unless, of course, you are also saying that GNU project was doomed from the beginning due to Stallman's ineloquence and lack of charisma.

"most authors don't even take the name as a question. Anyone can easily verify this." If it's easy, then verify it. If your excuse is that there is no published source, then do an interview or good survey of free (and open source) software authors/writers, and publish it in a reputable magazine, and then reference it here. Until then, "it's obvious" or "it's not even a matter of question" is not a valid argument for editors of an encyclopedia. novakyu (talk) 21:54, 29 January 2008 (UTC)

Wikipedia doesn't decide what's the "correct" name, the criteria for using a name or other are listed in WP:NAME. Give it a rest, even RMS admits that Linux is more widespread, I think his quote clears that if there was any doubt. -- AdrianTM (talk) 21:59, 29 January 2008 (UTC)

I don't have much time, so just a quick comment about rms's quote. As I argued, the quote was made more than a decade ago. If you want to say that more than a decade ago, "Linux" was an overwhelmingly popular term over "GNU/Linux", then yes, you are very well justified in saying that. On the other hand, if you want to say that "Linux" is overwhelmingly popular over "GNU/Linux" today, you should find a more recent essay/quote/article by rms or other free software advocate. You are free to disagree, but I hope to hear a more reasoned argument than "you are wrong, RMS said something a decade ago and it lasts forever and it proves my point irreputably." novakyu (talk) 23:06, 29 January 2008 (UTC)

I'm sorry but I don't have that much interest in the issue to look for more recent quotes (just like I'm not going to spend much effort to find quotes that water is wet), but I don't know why I should, has anything changed in the last 10 years in that respect? Maybe you should provide a quote that shows that GNU/Linux usage has significantly changed lately, as far as I'm concerned nothing significant has happened in that respect and since the quote is still on gnu.org site I assume it's still current, I'm sure they wouldn't keep obsolete info, especially when that's about the GNU/Linux name. -- AdrianTM (talk) 23:16, 29 January 2008 (UTC)

Besides the fact that the essay you are referring to was updated in 2007 according to the web page, you can find similar recent quotes by Richard (e.g. in this 2005 interview he also says that most people call it "Linux"). (Say what you like about RMS, but he doesn't have a problem acknowledging the reality that he can't always persuade people—he's been quoted several times as saying that he is a pessimist by nature, and usually expects to lose his battles, but fights them anyway.) In any case, as I point out below, the evidence is all around us; this is a little like requiring a reference and arguing about statistics for the claim that the Pacific Ocean holds more water than all the bathtubs in New York. —Steven G. Johnson (talk) 23:43, 29 January 2008 (UTC)

I'll answer to two points above for now (and I'll try doing a little research to answer the compelling argument below by Stevenj later, probably during the weekend when I have much more time):

First, AdrianTM, GNU (and FSF) website tends to keep "obsolete info" for historical purposes. For example, most of the "Boycott Amazon" website has not been edited, except, of course, the initial clarifying statement. Also, if you look at "Right to Read", actual body of the essay rarely changed over the decade (although since this one is a fiction ... I'm not sure if "obsolete" would ever apply to it), and instead, there's an author's note that gets updated every now and then. You shouldn't look at an essay on GNU or FSF site and decide, since it's still up, that it's current—no more than you can look at a newspaper in the library, and since it's still stored in the library, decide that it's current.

Of course, I don't have anything to do with the GNU or FSF website personally and what I say above is just speculation on my part. But regarding the fact that the "Linux and GNU" essay seems to have been edited in 2007 (and actually, 2008), my guess would be it's for the addendum. In any case, what I'm trying to say is that it seems very rare for FSF to edit articles surreptitiously so that every bit of it is absolutely current. Perhaps I'll e-mail the webmaster and ask him (or her) whether notes about GNU/Linux distributions that identify itself as GNU/Linux distribution (i.e. "Debian Linux" => "Debian GNU/Linux") should be added as a headway for this campaign, just to head off misunderstandings of this kind.

By the way, Stevenj, I would just like to point out that, unless I am having a huge reading comprehension problem, you seem to be misquoting rms. This is the exact quote from the article: "Most of the time, when people call something 'Linux', it's the GNU system with Linux as the kernel." (And the quotation marks below mark what I would imagine rms or anybody saying;) rms is not saying "when most people refer to the GNU system, they call it 'Linux'." He is, however, saying that "when most people say 'Linux', they mean the GNU system, with the Linux kernel." In other words, if you took the number of times people said "Linux" and meant the kernel only, and the number of times people said "Linux" and meant what some of us call "GNU/Linux", the instances of second case will be far more numerous. So, again, if I understood that quote correctly, this is not the recent quote (i.e. rms acknowledging that use of the term "Linux" for the free operating system using GNU tools is far more widespread than "GNU/Linux" today) that you are looking for.

I realize finding recent quote is not an easy job, but could it possibly be because "GNU/Linux" is not such a marginal term it used to be? novakyu (talk) 11:08, 31 January 2008 (UTC)

Regarding the FSF keeping old essays up for historical interest, that's true...but they also update the central essays periodically (this one was updated last year) to correct any factual changes in the situation. You're right that I misquoted RMS's 2005 essay, a fault of reading too quickly, but if you take your reading and combine it with the fact that "Linux" shows up orders of magnitude more often than "GNU/Linux" in literature/media searches, you have to accept that "Linux" is a far more popular term for the OS. (Unless you think that, contrary to what RMS accepts as fact, 99% of the time that people talk about "Linux" in the media they are talking about the kernel proper, which is stretching credulity.) I'm sure one can find a more explicit quote from this by RMS, but in any case we don't have to take RMS's word for it—as I said below, can you find a single mainstream media outlet that refers to the OS more often as "GNU/Linux", or where the usages of both terms are common enough that the issue is even remotely debatable? It might well be that "GNU/Linux" is more common than it was 10 years ago, but even a cursory look shows that the overwhelming preference for "Linux" (whether one agrees with this usage or not) hasn't changed. So far, you haven't given any empirical evidence to cast even the slightest doubt on this. —Steven G. Johnson (talk) 16:57, 31 January 2008 (UTC)

Let's look at a few mainstream sources. Eweek has a "Linux Watch" section covering the OS. [3]. Linux Journal ... um, enough said. Byte magazine has "Linux line" and "Serving with Linux" columns on the OS. Dr. Dobb's Journal has a Linux/Open-Source section [4]. The New York Times has a "Linux (Computer Operating System)" topic [5]. Ars Technica has a Linux.Ars section [6]. These are just a random selection of the first few places I tried. A search on the Factiva news database finds more than 100 times as many mentions of "Linux" as "GNU/Linux" in press sources. I'm sympathetic to the arguments for "GNU/Linux", but it seems hard to deny that "Linux" is the most widespread name if you look in any mainstream source. Can you name a single mainstream press source that refers to the operating system as "GNU/Linux" more often than as "Linux"? (The closest thing I can find is obviously partisan sources such as the GNU's Bulletin or Free Software Magazine, and neither of these is what one would normally call mainstream press.) —Steven G. Johnson (talk) 22:15, 29 January 2008 (UTC)

[1] ttp://homepage.ntlworld.com./jonathan.deboynepollard/FGA/dos-windows-boot-process.html

[2] ttp://www.google.com/trends?q=GNU%2FLinux%2C+Linux

[1]

[2]

Further reading?

Debian naming issue

"Mepis Linux"

Windows- originator of general public's concept of Operating System?

Reducing quotes

Re: removed initial sentence from pro-"Linux" argument