Talk:Tanenbaum–Torvalds debate

Proposed merge

I don't see how this article can be expanded and suggest merging it into the Architecture section of the Linux (kernel) article. The "debate" itself is part arguing over pros and cons of microkernels/monolithic kernels, which are described in their respective articles, part silly flamewarring. Memmke 10:49, 21 November 2006 (UTC)

I'll copypaste what I said in the previous AfD: It can be expanded with both views, that of the microkernel and the monolithic kernel, as well as an observation of currently existing kernels and thus the practical use of both design philosophies. The debate was revisited at a later date; we can cover that as well. I don't see how this cannot be expanded. Just give it eternity, and this article will eventually write itself, like every other article on Wikipedia. —msikma <user_talk:msikma> 10:56, 21 November 2006 (UTC)

But that is already covered in the articles microkernel and monolithic kernel, which are already linked to from the Architecture section of the Linux kernel article. Memmke 11:09, 21 November 2006 (UTC)

I think it's for the better good of Wikipedia to avoid redundancy and (using your wording) eternal stub articles. Memmke 11:12, 21 November 2006 (UTC)

The point of that wording was that stub articles do not remain stubs forever. I think you'd be surprised at how well something can be expanded simply by very thoroughly discussing the subject matter and explaining it in a way that's understandable for the average reader. Besides that, it is very relevant to the subject that things such as both sides of the argument get explained rather than just saying "Linus likes the monolithic kernel, go check out that article; Tanenbaum likes the microkernel, go read up on that to understand what he meant"; they have very distinct reasoning that would be both too esoteric and detailed for the respective kernel articles to convey. —msikma <user_talk:msikma> 11:53, 21 November 2006 (UTC)

And you are waiting for someone else to come along and write what you can't? Memmke 08:47, 22 November 2006 (UTC)

Please do not singlehandedly turn this article into a redirect without discussing it here first. Aecis ^{Dancing to electro-pop like a robot from 1984.} 10:32, 22 November 2006 (UTC)

I thought that that was what being bold was all about. Memmke 10:34, 22 November 2006 (UTC)

WP:BOLD also says "be bold in updating pages" does not mean that you should make large changes or deletions to long articles on complex, controversial subjects with long histories, such as the Israeli-Palestinian conflict or abortion, without carefully looking at your edit. By turning an article into a redirect, you are basically blanking that article, and that should be done with caution. I'm not saying you shouldn't be bold, please do, but sometimes a little caution might be needed. Aecis ^{Dancing to electro-pop like a robot from 1984.} 10:44, 22 November 2006 (UTC)

I see your point. I suppose we'll wait and see then. Memmke 10:52, 22 November 2006 (UTC)

Of course I'm waiting around for someone else to write this article. I can't write every article on this site. I have other priorities. But just because I can't edit this page right away, does that mean we should just not have the article at all? I don't see how this makes any sense. —msikma <user_talk:msikma> 12:56, 22 November 2006 (UTC) PS: don't leave silly messages on my talk page; you're the new kid on the block, not me. See WP:POINT.

I'm not saying that you should write every article, just that it may be a good idea for you to continue what you initiated with the ideas you've suggested. Oh, and that message, I just had to. Memmke 16:53, 22 November 2006 (UTC)

I certainly intend to begin writing this article at some point, but I'm not willing to do it just yet. That in itself is not a reason for the article to be merged. The amount (or lack) of development an article has gone through is not an argument for any action of the sort. —msikma <user_talk:msikma> 21:03, 22 November 2006 (UTC)

I agree with the merge: this debate only deserves a passing mention in Linux kernel or somewhere else. It's really a minor event in the history of Linux, and of zero importance outside the world of OS'es. It's probably not even worth to mention in Microkernel. Tizio 19:26, 22 November 2006 (UTC)

Perhaps to you. The general consensus in both AfDs was that this article should under no circumstances be deleted; the reasons given pertained to its notability. This is certainly notable enough for inclusion, and on top of that, it's sufficiently notable to be an article (rather than a section of an article). This was, in fact, widely reported, both on Linux sites as well as in the news. Same when the debate "heated up again". I usually prefer merged articles (as I prefer long, monolithic articles rather than fragmented articles) but I don't see any reason for a merger here. —msikma <user_talk:msikma> 21:00, 22 November 2006 (UTC)

The decision was to keep the content, not to keep it as a separate article; most of the people suggesting keep did not rule out a merge. "Could be expanded" is not a reason for keeping an article separated, since it can be expanded even if merged. Tizio 21:23, 22 November 2006 (UTC)

Hardly anybody in either AfDs suggested a merge. Actually, I see only one clear "merge" vote in the first one and zero in the second. Whether the people who voted for the article to stay "did not rule out a merge" is something that can't be said so easily. As far as I can see, it was put up for deletion, but was found to be notable enough to stay. If the people who voted really wanted it merged, they'd have voted with "merge" or "keep and merge" instead. Also, "could be expanded" is, as you put it, not a reason to keep an article separate, but it certainly also isn't a reason to merge it. —msikma <user_talk:msikma> 22:20, 22 November 2006 (UTC)

Hardly anybody? Me and Sandstein wanted a merge, Tractorkingsfan and Georgewilliamherbert didn't mind it either. And you can't use the second AfD as an argument for its notability or whatsoever as it didn't debate that, it merely speedily reinforced the prior decision. Memmke 08:25, 23 November 2006 (UTC)

What I am pointing out is that most of the people on AfD simply did not comment on merge, neither to support nor to oppose. Therefore, it is up to us, in this talk page, to make this decision.

Regarding the issue of several sources reporting about this debate: I am not commenting about deleting this article; however, having it as a separate article gives it an undue weight. In my view, this is kind of a piece of gossip; it is something that spreads as a sort of weird news, such as "heard of this? Tanenbaum once told he would have failed Linus Torvalds for proposing Linux". The fact that it was widely reported is not important: we do not have an article on 1996 Oprah Winfrey weight loss, even if that was widely reported too. Tizio 11:21, 23 November 2006 (UTC)

Perhaps it is gossip to you. To me it is not. I find 75,800 relevant search results, discussion of this subject in notable publications (for example, the O'Reilly book Open Sources: Voices from the Open Source Revolution includes the first instance of the discussion completely; there are also publications on important websites such as by Tanenbaum himself for the Vrije Universiteit Amsterdam, or Slashdot and many other major news sites) to be more than plenty of proof that this isn't just gossip. It was widely reported on and discussed in detail by sound and relevant publications, which should account for the fact that it isn't just sensationalist journalism, as would be the case with 1996 Oprah Winfrey weight loss. —msikma <user_talk:msikma> 11:48, 23 November 2006 (UTC)

Major media sometimes report about frivoluos things; that Oprah thing was probably also reported in media we generally consider reliable as of, for example, politics. To me, the litmus test of notability is how influential a subject has been. I see you have many google hits and also some reliable references; however, I do not see how this debate has, for example, affected the subsequent development of Linux (or of any other OS), and I also do not see how it affected anything outside of the Linux world. Tizio 12:36, 23 November 2006 (UTC)

Stephen Colbert at the 2006 White House Correspondents' Association Dinner also did not affect the White House or the world of comedy; it was just "reported about". Since when it is a requirement that there is a significant influence on the outside world for something to be deemed "notable"? —msikma <user_talk:msikma> 12:48, 23 November 2006 (UTC)

What I said of this paper applies to that as well: it deserves some mention in the comedian's article, but not an article by itself (let alone a featured article). I however also noticed that that article claims the "event" to be an important part of politics of U.S. in 2006, while no claim of importance is given in this article. Tizio 13:11, 23 November 2006 (UTC)

Again, that's just what you think. You don't think that the article I linked to should be a separate article? Then propose for it to be merged, and you'll see that the community disagrees. That's because articles like that one deserve to exist; as is the case with this one. It's also blatantly ridiculous that you don't think it should be a featured article. That's mixing up priorities again; the quality of an article has absolutely nothing to do with whether it has the right to exist. Articles that aren't well-developed should not be deleted or moved out of sight for that reason, as otherwise we'd not have any well-written articles. —msikma <user_talk:msikma> 13:21, 23 November 2006 (UTC)

Yes, that's what I think. Were you expecting me to post something I do not think? I am not wasting my time on that other article, given that I am already unable to get consensus on merging this one. Regarding the featured status, I already know the policy of featured articles; however, I think it's bad that we present such an article as "one of our best ones". And no, I am not proposing the merge because this article is under-developed, as I have said above; rather, it's because it should not be developed further. Tizio 14:02, 23 November 2006 (UTC)

By "just what you think", I intended to imply that it is the opinion of an individual and not a general consensus. The general rationale used to merge or delete an article isn't what you mentioned and simply does not apply to this one. —msikma <user_talk:msikma> 14:26, 23 November 2006 (UTC)

Okay, avoiding that whole trainwreck of a discussion above...

There has been a proposal to merge, which I will assume to have been in good faith. An AfD keep does not preclude or discourage a merge; it is simply consensus that the article should not be deleted. The major argument against the merge appears to be the contention that the article is expandable into a full-length article that can stand on its own.

It seems to me, then, that the solution is simple. Supporters of keeping the article separate should work to expand it and prove their point. Supporters of merging the article, meanwhile, should refrain from interfering with such efforts. Try to agree on a reasonable amount of time to allow for some expansion to take place -- I would recommend one month -- and review the article's situation at that time. If no reliable sources have been found that would permit expansion, the argument to merge is that much stronger. If the article has grown, then the argument to merge is clearly moot. Either way, more is accomplished than trading "is not"/"is too"-style dialogue or engaging in edit wars. Shimeru 20:44, 23 November 2006 (UTC)

I disagree with that. This article is clearly able to stand on its both feet. A tiny bit of research into the topic yields plenty of material that may be added to the article. I am not going to hurry up and hastily write up this article as though forced by these people. This article should remain exactly as it is, a needy stub, until someone begins writing this article. There are literally thousands of other needy stubs that are legitimately separate articles, and I don't see how this one is different. —msikma <user_talk:msikma> 20:53, 23 November 2006 (UTC)

That's fine, but it obviously is not "clearly" able to stand on its own if people are proposing deletions and merges. An article should not remain a stub forever, and asserting that "there are lots of sources" does little good if those sources are not used. Even if this is an important topic, one must ask whether the article itself is of any usefulness to those interested in the topic. Right now, it provides little information. In such a case, a merge may be the best approach -- the article can be split again at a later date, when interested parties have the time to expand it and make something more informative of it. On the other hand, existing as a stub creates a sort of open invitation for those who are interested to make further contributions. But if those contributions are not made, for how long should it remain a stub? A month? Six months? A year? Forever? Shimeru 21:17, 23 November 2006 (UTC)

The amount of work that has gone into an article does not in any way correlate to which actions should be taken. An article isn't merged or even deleted for being undeveloped. That's blatantly ridiculous. Articles will "write themselves" if you give them the time. And yes, this article will remain a stub for forever if nobody wants to write it; that's just the way it always has been on Wikipedia. But the thing is that the article won't be a stub forever, because if given time, someone will come around and start working on it. I'm confident that someone will. If we merged all tiny articles for the simple reason of them being small, we would have much less articles than we have today. I don't see how anyone can say that this article cannot exist; it's a very notable subject that's been published on widely, both on the Internet and in print, and has been the focal point of discussion more than once. Saying that someone must start working on this article because otherwise it shouldn't exist is more or less blackmail to those who acknowledge that it should exist. —msikma <user_talk:msikma> 07:11, 24 November 2006 (UTC)

Just because there's a lot of abandoned stubs out there doesn't mean we don't have to care about this one, and about the number of articles, I think we should strive for quality of articles, not quantity. Why this article may not be like the "rest" of the stubs, or why it may not write itself eventually, is IMHO because of the little possibility of useful expansion. And at its present state, it just disrupts readers of the articles it is linked to. Memmke 08:50, 24 November 2006 (UTC)

We should indeed strive for quality of articles. I wholeheartedly agree with you on that. I disagree with your saying that there is little possibility of expansion here. There are plenty of things (I've mentioned it before, and it has also been mentioned in the AfDs) that this could be expanded with that are clearly outside of the scope of, for example, the Linux kernel article(s). As for "articles writing itself", this is fundamental to Wikipedia as a whole. If this weren't true, there would not be any good articles. Articles are always written at some point, even if it is about a niche subject. —msikma <user_talk:msikma> 11:29, 24 November 2006 (UTC)

Articles are merged for being undeveloped, actually. Perhaps not as often as they ought to be, but not infrequently. On a side note, you don't need to convince me as to the notability of the debate; I was a Unix sysadmin at the time, followed it somewhat closely, and was even tangentially involved. More importantly, the community consensus was to keep the information. What I'm trying to get across is, if no further information than this is available, it may be more useful to interested readers to merge the information into a related articles until such time as someone does see fit to expand it. I'm not certain why this is bad because it reduces the number of articles -- it doesn't reduce the amount of actual information, just the amount of searching and link-clicking a reader needs to do to find that information. Now, bearing that in mind, you say there's a great deal of information about the topic available, and that the article could be expanded. But it appears that you aren't prepared to perform that expansion, and that you aren't aware of anyone else who's planning to perform that expansion. Yet you obviously feel it's important that this article remain separate for whatever amount of time it takes for someone to get around to expanding it. Why? (I'm not asking this adversarially; I'm just trying to understand your reasoning.) Shimeru 10:45, 24 November 2006 (UTC)

Typically, underdeveloped articles are merged due to the impossibility of expansion, because the subject matter may be narrow. Such a merger may also be prevented by the fact that the article is really outside of the scope of the article it could be merged with. While I agree that this article may be seen as a grey area concerning these two things, I still think that stands apart from typical Linux kernel (or monolithic kernel or microkernel) discussions enough to be on its own, and that it's also sufficiently diverse to be expanded to a large article, which are the two technical reasons for me opposing this merger. As for people expanding this article: does it really matter whether an article is expanded straight away or in the future? We never know when someone is going to come in and expand it. It may be today, it may be tomorrow. It takes just one dedicated editor, and it could be brought up to A-status in a short amount of time in the best case. All articles are left on this site with the hope that it will be edited to be more useful, and this article is no exception. I've got other priorities now, as do others who could potentially edit this article (seen as how there have been few contributions so far), but that will most definitely not be the case in the future. I've already stated that I want to edit this article — just not right now — and that's enough proof to show the system works. All articles have been underdeveloped at some point. I don't think it should be a reason for an article to be merged, even if I'm generally a fan of large and broad articles instead of small, fragmented articles. —msikma <user_talk:msikma> 11:29, 24 November 2006 (UTC)

Okay, thank you. If I'm correct in my assumptions, though, it seems neither you nor the other principally-involved editor, Memmke, are prepared to shift from your respective positions or compromise on the matter. Please correct me if I'm mistaken. Shimeru 21:08, 24 November 2006 (UTC)

I'm not willing to compromise because I believe that I'm right. :) I still hope to convince Memmke and the others that this article should be on its own, but it doesn't matter much anymore, I believe, since a consensus on merge will likely not be reached anytime soon. function msikma(user:UserPage, talk:TalkPage):Void 07:10, 27 November 2006 (UTC)

Work does seem to be progressing nicely; I feel my main concern above has been addressed. Well done on the sourcing. Shimeru 20:37, 26 November 2006 (UTC)

Thanks. I think that I'll at least get this article into B-quality before returning to other articles. Preferably GA-quality. In a few edits, the scope of the discussion will become visible, and by then at least it'll be out of start quality. function msikma(user:UserPage, talk:TalkPage):Void 07:45, 27 November 2006 (UTC)

I must admit that I'm also quite satisfied. Nice work msikma. Memmke 08:28, 27 November 2006 (UTC)

Coverage

Some of the things I think the article should cover:

• That the debate widely has been regarded to be, to some degree, a flamewar. (Although an informative one.)
• Tanenbaum's goals with his Minix; such as portability, to be a pedagogical teaching aid, and the possibility for students to run it on cheap hardware.
• The great hopes for the GNU project's HURD kernel many of the participants seem to (have) hold.
  • I think that we should cover this among the other things that the participants seem to believe the future will hold. They also wrongfully predicted the downfall of x86. function msikma(user:UserPage, talk:TalkPage):Void 13:12, 28 November 2006 (UTC)

Memmke 09:24, 28 November 2006 (UTC)

Tentative listing of participants:

• Michael Haardt
• Peter MacDonald - Early Linux kernel developer and creator of one of the first distributions; Softlanding Linux System
• David Megginson
• David Miller - Linux kernel developer
• Peter da Silva
• Andrew S. Tanenbaum - Creator of Minix
• Ken Thompson - Co-founder of Unix
  • Incorrect--same name, different person. The Ken Thompson of GTRI who participated in the original debate was not the Ken Thompson of Unix pioneer fame. The Unix pioneer was still at Bell Labs at the time of the debate, and was never associated with GTRI. The Ken Thompson of GTRI sometimes posted using the name W. Kenneth Thompson, while the Unix pioneer's full name is Kenneth Lane Thompson.
• Richard Tobin
• Linus Torvalds - Creator of Linux
• Theodore Ts'o - The first North American Linux kernel developer, who were also the first to set up an ftp server for Linux in the North America

Memmke 10:09, 28 November 2006 (UTC)

Excellent. I'm glad to see you're willing to help out on this article as well. function msikma(user:UserPage, talk:TalkPage):Void 13:12, 28 November 2006 (UTC)

Yeah, I thought I'd be a little more constructive. Memmke 08:37, 29 November 2006 (UTC)

Reduced flame war

I have changed

Due to the strong tone used in the newsgroup posts, the debate has widely been recognized as a flame war, a deliberately hostile exchange of messages, between the two camps (of Linux and MINIX, or alternatively, of monolithic kernel enthusiasts and microkernel enthusiasts) and has been described as such in various publications.[3] Torvalds himself also acknowledged this in his first newsgroup post about the issue, stating (verbatim) “I'd like to be able to just 'ignore the bait', but ... Time for some serious flamefesting!”[4]

The debate has sometimes been considered a flame war.

My reasoning is: The debate is in it self valid, and using the term flame war can give the wrong impression (at least when a disproportionate part of the article is spent on the subject). Flame wars are associated with personal conflicts; yet, the personal relation between Torwalds and Tanenbaum seems to be good. The cited source use "flame war" in quotes, which could imply that it uses the term in a very wide (or even joking) sense (I have not followed it up). The quote by Torwalds comes from the very beginning of the discussion and is speculative, and thus says nothing about the overall debate. If a flame war as taken place between "secondary debaters", this should not be mentioned in this manner, because it detracts from the debate between the primary debaters. There is a difference between "heated debate" and "flame war" where, IMO, in dubio pro reo judgement should be made.

I will not object, if a later expansion of the topic takes place. However, I strongly suggest that a stronger foundation is given for the claims; further, that a discussion of flame wars, if re-introduced, is put in a section of its own and not in the introduction.

88.77.153.201 (talk) 02:30, 15 June 2009 (UTC)

Jonathan Shapiro

"This prompted Jonathan Shapiro to respond"

Who is Jonathan Shapiro? There should either be a link here or a brief explanation. Will Faught (talk) 04:53, 17 January 2013 (UTC)