Talk:Plan 9 from Bell Labs/Archive 1

Better term?
Is there a better term we can use describing Plan 9 rather than Unix-like? If you put a *nix user on a Plan 9 box, he or she would not be able to perform even basic functions. Same goes for network programming. And application programming. This is just a nit-pick on one word. Plan 9 is really nothing like a modern (or even an old) *nix variant. Conceptually, it is an expansion of several concepts in Unix - but I wouldn't call it Unix-like. --ElBenevolente 10:11, 23 May 2004 (UTC)


 * Indeed they are really nothing alike. They have some similiarities, but then again so do all systems. "Those who don't understand Unix are condemned to repeat it. Badly." Plan9 has a little more similiarity than many, but I think the article should either clearly point out that they are not alike or even related or avoid mentioning Unix for anything more than some common history. --Anon.


 * Are you claiming that the developers of Plan 9 didn't understand Unix? I hope I misunderstood you. Plan 9 and Unix share the same core principles and philosophy, both were developed by the same people, at the same place; the main differences between Unix and Plan 9 are where people that did not understand Unix extended it and added "features" that didn't fit the Unix philosphy. --Lost Goblin 00:39, August 14, 2005 (UTC)

9p page?
Hi! In looking at the 9P page, I couldn't quite figure out what made it special. The description for 9P on this page does a better job. I almost copied some info over myself, but since I know nothing more than what I've read here, I was hoping someone more familiar with Plan 9 could take a swing at it. Thanks! --William Pietri 04:58, 1 November 2005 (UTC)

Licensing details
Could someone ammend the article to explain whether the currentl Plan 9 license is GPL-compatible or not and if not, why? --Ssokolow 03:43, 17 January 2006 (UTC)

Relevance of popularity
At 2006-03-25 13:31:30, Gnetwerker performed an edit seeming to rise from confusion over what it means to claim Plan 9 is UNIX's successor. Subsequently, several attempts have been made at resolving the legitimate initial confusion, but Gnetwerker seems intent on maintaining some mention of relative popularity in the introduction. This is not relevant here. Having resolved the real issue, any mention of comparative popularity belongs in a section on place in the market, influence on the industry or similar. Gnetwerker's agenda should not control the page. Comments from other editors (including Gnetwerker, of course)? --AnthonySorace 18:54, 25 March 2006 (UTC)


 * "Popularity" among who? as measured by who based on what parameters? I think this is offtopic and extremely subjective and doesn't belong there. Plan 9 is certainly much more popular than Unix 10th Edition which was arguably the last version of Unix. --Lost Goblin 18:57, 25 March 2006 (UTC)


 * Not only that. I think the people who made Unix get to chose what its successor is. It would be very rude to the people who made Unix and Plan 9 to take away this right or claim it from them. --ems 19:08, 25 March 2006 (UTC)


 * Most people would assume that Linux is the successor to Unix. Clearly Linux does not have Bell Labs heritage, but it is the Unix-like operating system most widely used, so for that definition of "successor", it is clearly the winner.  The most popular version of actually Unix-derived OSs at the moment would probably be Solaris.  Plan 9 was an interesting research experiment, but for a wide variety of reasons, has not succeeded commercially or as free software.  You have also created the category of "Unix successors", and it would appear appropriate to add Solaris, Linux, and perhaps others into that category.  Observe that WP:LEAD says "The lead should briefly summarize the most important points covered in an article in such a way that it could stand on its own as a concise version of the article."  I would think that the most important point from the (nonexistant) "influence" section.


 * In terms of relevance, when an "improvement" to a very popular system fails to catch on, I would say that this is relevant. --Gnetwerker 19:26, 25 March 2006 (UTC)


 * "fails to catch on" among who? and in what way? Wikipedia is certainly not about "winners"(what is the competition anyway?) or subjective perceptions of popularity, it is about factual and relevant information. As for successor, I don't know what definition you are using, but the one I know is this and Plan 9 follows it perfectly as Unix successor:

successor n 1: a person who follows next in order; "he was President          Lincoln's successor" [syn: {replacement}] 2: a thing or person that immediately replaces something or        someone 3: a person who inherits some title or office [syn: {heir}]


 * Bell Labs created Unix, and they got to pick its successor. Also note that I know many Unix people that would consider Linux or Solaris very inferior to Unix 10th Edition and that probably would not even consider them worthy of the Unix denomination, much less its successors, to replace something you have to at least fullfill the same purpose, and Plan 9 fullfills much more closely the purpose of Unix than any other system.


 * The whole article needs lots of cleanup, maybe mentioning that Plan 9 currently has less users than other systems might make sense in some context or maybe not, but there are many much more relevant issues to be covered. Also any assessment of popularity should be careful not to fall into "original research". --Lost Goblin 19:53, 25 March 2006 (UTC)


 * It is not true that a comment on the relative user base is the most relevant summary of the (pending) "influence" section. For particular examples, see /proc in most modern UNIX derivatives, especially Linux, and v9fs in Linux (and, earlier, in now-outdated BSDs). You're confusing the primary goal of a commercial operating system (gain market share) with those of a research operating system (explore interesting ways of solving problems).
 * The intent here is certainly not to imply that Plan 9 has replaced UNIX in industry, nor that it's a sufficient drop-in replacement functionally, generally. But it is true that it was the successor to v10 unix in nearly every way, the the successor for UNIX generally for Bell Labs' operating system research. The idea that Linux is the successor in any of these terms is simply factually false.
 * I agree the relative popularity has a place in the article, but it's not in the summary. This seems to be the community consensus. --AnthonySorace 20:17, 25 March 2006 (UTC)


 * rc and sam are also very popular in many Unix circles, not to mention UTF-8, or that Xrender is an almost exact copy the Plan 9 draw interface. I think assessing "popularity" is a very slippery ground that can easily lead to "original research" and POV. --Lost Goblin 20:23, 25 March 2006 (UTC)


 * Perhaps it is the consensus of a few Plan 9 promoters, but I don't know about "the community". Perhaps and RfC is in order? -- Gnetwerker 20:24, 25 March 2006 (UTC)

If you like to waste your time over such minutiae go ahead, I certainly have more important things to worry about. --Lost Goblin 20:41, 25 March 2006 (UTC)


 * Note that, for example, the Solaris and AIX articles make no mention of popularity in their intros. Presumably that speaks to the broader community's opinion on the relative relevance of generalized assessments of popularity of the individual systems. --AnthonySorace 20:54, 25 March 2006 (UTC)

Move
The correct name for the operating system is Plan 9 from Bell Labs. --ems 16:17, 25 March 2006 (UTC)

(6/0/0) voting closed outcome: moved.

Support

 * 1) Support - voting sucks. --ems 16:17, 25 March 2006 (UTC)
 * 2) Support - "Plan 9 from Bell Labs" has always been the proper name, "Plan 9" and "plan9" are acceptable colloquially but are not the full name, also "Plan 9" is ambiguous as can be seen in the Plan 9 disambiguation page Lost Goblin 16:27, 25 March 2006 (UTC)
 * 3) Support - with redirect, since many people will come looking for the OS called "Plan 9". AnthonySorace 17:14, 25 March 2006 (UTC)
 * 4) Support - I had thought that the article's name was a little odd before. Benn Newman 21:20, 25 March 2006 (UTC)
 * 5) Support. Funnier names == teh goodness. --maru (talk) contribs 00:25, 26 March 2006 (UTC)
 * 6) Support - With the name of the OS being "Plan 9 from Bell Labs", the wikipedia article name should reflect that. --pippijn 12:50, 26 March 2006 (UTC)

Impact
Someone added these lines (I've cleaned up the grammar, but the sense is the same):


 * "The growing interest in distributed systems and grid computing has brought again attention to the distributed capabilities of Plan 9 and 9P in particular with numerous client and server implementations for a wide variety of systems. Plan 9 proponents say that the problems hindering its adoption have been solved, and its original goals as a distributed system, development environment, and research platform have been met, and that it enjoys moderate but growing popularity. Inferno through its hosted capabilities has often been a vehicle and bridge to bring Plan 9 technologies to other systems particularly as part of heterogeneous computing grids."

I know of no reliable secondary sources that support any of this. Please supply them or we should remove or reword. --Gnetwerker 22:59, 25 March 2006 (UTC)

EMS, if you have a problem with "out of context", then add context, don't delete the quote. -- Gnetwerker 06:44, 26 March 2006 (UTC)


 * I never deleted it. I commented it out. Wikipedia not a source of misinfomation. You took half of his sentence and changed into something totally different. You changed something he refuted into something holds. Until this is fixed, the section should be commented out. ems 07:37, 26 March 2006 (UTC)


 * You appear to want to create a fan page for Plan9. That is something Wikipedia is not. You need to stop trying to spin everything about Plan9 in a positive light. Anyone who did not have a POV on the subject would realize that the Raymond quote was used exactly as intended, merely shortened. However, if you want to lengthen it, fine. Just don't mess with the central point: Plan9 lags far, far behind other Unix descendants. -- Gnetwerker 16:59, 26 March 2006 (UTC)


 * Do you read what you write? If "Plan9 lags far, far behind other Unix descendants." is not POV I don't know what could be. This is becoming ridiculous, do you have any actual informative content to provide, or do you just intend to continuously push your agenda of showing Plan 9 as a complete failure? (That would require at least a knowledge of what the Plan 9 goals are, which you seem to lack even when they are partially documented in this page itself) You talk much about facts, but I aside from your selective quoting of ESR (whom I plan to contact to clarify the quote, as I think has been taken out of context and is being misused) you have yet to produce anything that contributes any value to the page except misinformed and unsustained rants about things that are hardly relevant. If you have any valuable and non-speculative information to contribute, please do so, but please don't systematically interfere with what others are trying to contribute, criticism is always helpful and appreciated, but way enough time has been wasted in this pointless and childish argument. -- Lost Goblin 18:03, 26 March 2006 (UTC)


 * Would you like Eric's phone number? I have been a Unix developer since the mid-1970s, and booted Plan9 when you were probably still using training wheels. Plan9's market status is a matter of record, not my opinion. The problem here is that you and Ems want this to be a Plan9 fan page, not an honest encyclopedia article. Maybe you can get a number from Vita Nuova about how many units were sold last year? -- Gnetwerker 18:09, 26 March 2006 (UTC)
 * If it is a matter of record, provide references. And BTW, Plan 9 is not primarily a commercial product, and has never been one. -- Lost Goblin 18:18, 26 March 2006 (UTC)
 * I also note on your personal web page (linked from your WP user page) the statement: "I am a fan of Plan 9 and Inferno trying to spread their ideas and technologies back to the Unix world." - Wikipedia is not the place to do that. -- Gnetwerker 01:40, 27 March 2006 (UTC)


 * BTW, it has been discussed already that comments about things like "popularity" and "success" (not to talk of "lag behind" and other demeaning terms) are very subjective and a very slippery slope towards POV, still you keep pushing for such comments to be included. If you disagree with what had been discussed in the talk page before about this, please comment on it before persisting on such changes. Ems2 changes have been at times not careful enough and might have contaminated POV bits on those same subjects, lets try to keep the page clean of it and not waste more time with matters that are so subjective and of so little value. -- Lost Goblin 18:18, 26 March 2006 (UTC)

"For the full year, Linux server revenue was $5.7 billion" ; "Linux owns more than 5 percent market share for the desktop, far less than its server share."; "IDC said, Linux shipments will grow at a rate of 28 percent, from 1.3 million in 1999 to 4.7 million in 2004." -- Gnetwerker 18:43, 26 March 2006 (UTC)


 * Now do the same for each of the *BSDs, BeOS, Gnu/Hurd, etc, etc, etc. This is pointless. --Lost Goblin 18:48, 26 March 2006 (UTC)


 * Can we just leave it at Plan 9 is a hobbyist/research OS? There are very few commercial installations, and generally that relegates an OS to hobbyist (if used by technically proficient users) or research (if prevalent primarily in academic settings) status. --maru (talk) contribs 19:16, 26 March 2006 (UTC)


 * I think "hobbyist" OS is an accurate term. I am uncertain about "research". I am not aware of any serious research papers about Plan9 since the 1990s 2002. --Gnetwerker 19:34, 26 March 2006 (UTC)


 * Well, those sort of adjectives don't necessarily have to be present tense, I don't think. --maru (talk) contribs 19:45, 26 March 2006 (UTC)


 * You have not looked very carefully, take a look at the papers section of the Plan 9 wiki for some examples (there are more even if a bit harder to come by), things as important as 9P2000, secstore, factotum, venti(wont Best Paper Award at the first FAST conference) and fossil have all been released in the 2000s, and you can be sure more is on its way. Just because no one is paying attention doesn't mean that things are not getting done, also despite your claims, Plan 9 developers are more interested in writing code than writing papers --Lost Goblin 21:57, 26 March 2006 (UTC)


 * Feel free to post links to these -- Gnetwerker 22:00, 26 March 2006 (UTC)
 * I can't see why should we link every Plan 9 related paper from here, the relevant ones are already linked to from their respective articles (or should be, some articles are missing or need work, we are working on it, you are welcome to help) --Lost Goblin 22:06, 26 March 2006 (UTC)


 * Methinks thou doth protest too much. The last findable Plan 9 related paper was in 2002. If not true, please correct with citation. Plan 9 hasn't made an appearance at SOSP in a long, long time. -- Gnetwerker 22:11, 26 March 2006 (UTC)


 * Just a few examples, there are more but I can't be bothered to dig them up: Providing Asynchronous File I/O for the Plan 9 Operating System(2004), Real-Time in Plan 9 by Pierre G. Jansen and Sape Mullender  (2003) and a whole set of Plan B papers have been released in 2005 and 2006  (Plan B is a friendly fork or branch of Plan 9 and most of the changes are in process of being integrated back into Plan 9). And by the way your assumption that one has to write papers to do research is rather sad, unlike other research systems, Plan 9 is also used to get work done so most of the research is applied research; again, unlike most research systems more time is spent writing code than worrying about publishing papers. --Lost Goblin 22:42, 26 March 2006 (UTC)


 * FYI, the asynch I/O link is dead. --maru (talk) contribs 22:59, 26 March 2006 (UTC)


 * It works, but contains a colon which wikipedia doesn't like, let me try another way, yup, on its own it works, strange. I guess URL-encoding would work too. --Lost Goblin 23:04, 26 March 2006 (UTC)


 * Huh. That's pretty odd. But I vaguely remember running into a similar problem once, which I somehow solved through devious usage of nowiki tags.
 * I also find it interesting that it's an MIT bachelor/master's degree thesis; I seem to keep on running into useful such theses (be they on Plan 9, the AI Winter or what have you.) --maru (talk) contribs 23:42, 26 March 2006 (UTC)


 * Yes, so what is your point? Windows made billions more, so what. If you see Linus Torvalds announcement of Linux you will see he called it a Unix CLONE, not a successor. ems 15:41, 27 March 2006 (UTC)

ESR quote

 * Please, please get that Raymond filth out of there. I tried but somebody didn't like my replacement (please look at what I wrote!). He is an insipid ignoramus, hardly qualified to comment on Plan 9- let alone in an encyclopedia- in his warped and limited Linux-oriented view of 'Unix History'.) Just look at consumer feedback to his book on Amazon.com.

My only wish is that someone less in awe with the contents of his own field of vision, and with greater depth and objectivity (not to mention humility) had the opportunity to write this book.

Quality of discussion is varied as expected; Raymond is not quite the UNIX expert he thinks he is. In places, Raymond's tone encourages one to throw the book at the nearest wall and go out just to get some fresh air; He is condescending, hectoring, lecturing, and sometimes just misleading. Alas, I will still recommend it as worth reading (check your local library) with a nice grain of salt This book isn't about unix programming, its about the authors idea of linux programming. I will teach you bad habits and how to make non-portable linuxware. Hate to be the one to burst the proverbial bubble, but ESR's theory and ideology is ridiculously flawed with misappropriated valuations. This is, yet another advocacy campaign for gnu/Linux, mixed in with a UNIX context. Given the purpose of the book, it's a fair assessment to label the burdening bias as filler and firewood: filler for those who really just wanted an explanation of the single-purpose, POSIX illustrated, hows and whys of UNIX programming philosophy; and firewood for those who tend to buy into slanderous hype at the whim of suggestion.

Questions: What is the title of the book? What programmatic philosophies are portrayed? How many unices are open source? Of those, how many subscribe to the same opensource mentality as gnu/Linux? Answer: zilch. Question: Then what is the relevance of such topics to the objectives of the book, as depicted in the title? Answer: fudd-ala-mode.


 * So please, expose this quack, and get his droppings off Bell Lab's beautiful work.--216.101.25.243 06:01, 20 February 2007 (UTC)


 * What you wrote was an editorial, bordering on a rant. It's not appropriate for Wikipedia. Haikupoet 17:09, 20 February 2007 (UTC)


 * Ditto. I think ESR makes a valid point about why Plan9 hasn't taken over the world yet, which is as some editors previously mentioned, a useful antidote to Plan9 fanboyism; the comments on network effects, Worse is Better, and so on are all valuable. ESR is not good at coding, but his writing is not too bad. --Gwern (contribs) 21:34 20 February 2007 (GMT)


 * Wikipedia, facilitated by Google search, propagates an ignorant and tasteless middle-ground to topical neophytes. It irks me to think that Raymond, with no authority or taste on the matter, will be the one to turn away Plan 9 newcomers.  Then again, it's fitting for such a quote to be inserted by 'editors' that too enjoy belching their vague notion of a subtle topic!--216.101.25.243 01:49, 21 February 2007 (UTC)


 * Dude, cool it on the purple prose. You can make your point without sounding like someone doing a bad Captain Picard impression. The fact is, that rightly or wrongly (and most people would say rightly, though not uncontroversially) Eric Raymond is a recognized expert in the field. Certainly opposing views are always welcome in the interest of balance, as long as you can source them (which you didn't your first time around). But you're coming at this with a cause and a gigantic pitcher of Kool-Aid, which is not a good standpoint for writing an encyclopedia article. Also, would it kill you to get yourself a real login name? Haikupoet 02:18, 21 February 2007 (UTC)

Improper talk page cleanup
Apologies for the talk page deletions I made, I didn't know that was not the recommended way to clean a talk page. Thanks maru for correcting my mistake and showing me the errors of my ways, I will try to be more careful in the future. --Lost Goblin 01:43, 26 March 2006 (UTC)


 * Just next time remember how to archive a talk page. --maru (talk) contribs 01:48, 26 March 2006 (UTC)

Tense
Many parts of this article are written in past tense, for example "Key to supporting the network transparency of Plan 9 was a new low-level networking protocol known as 9P." Other parts use the present tense, I think that when describing the current system the present tense should be used, the past tense should be used for things that have been replaced (eg., Alef) If no one objects I will try to go over the current text and make it consistent, if someone can look over my shoulder it would be very appreciated because my English grammar is far from perfect. -- Lost Goblin 22:57, 26 March 2006 (UTC)


 * Such is the strength of Wikipedia -- there is always someone looking over your shoulder. I would recommend that past tense be used for retrospective comments about the goals and motivations of the developers, and current tense be used only for things that can be said to be true today, and not reflecting on original motivations. -- Gnetwerker 01:32, 27 March 2006 (UTC)


 * Personally, I would say that if you are describing a installation neutrally ("Plan 9 uses a networking system peculiar to itself to seamlessly network, called 9P..."), it would be present tense as any installation uses it (for example, one could write of VMS that it uses, present tense, a versioning system for files, even though it's a dead OS since any running installation does exactly that), but that first example in the past tense should be in the past tense- as in order to achieve the desired effects, they had to write 9P. It was key. --maru  (talk)  contribs 02:01, 27 March 2006 (UTC)

New intro proposal
I propose the intro to be changed to: Plan 9 from Bell Labs is a distributed operating system, originally developed at Bell Laboratories Computing Science Research Center, who also developed Unix. By the mid 80s, the trend in computing moved away from large time-sharing computers with multiple terminals towards networks of smaller personal computers. Bell Labs thought that UNIX -- itself being an old time-sharing operating system -- had trouble of adapting to these new idea born after it, especially in the realm of networking, graphics, and networked personal computers. Plan 9 began in the late 1980's as an attempt to answer the new trend of computing. Thus starting a new catch phrase, build a UNIX out of a lot of little systems, not a system out of a lot of little UNIXes. 

Plan 9 was started as an internal project to start Unix from the ground up with modern ideas of computing. By 1989, Plan 9 was used by the people working at Bell Labs as their operating system of choice. In 1992. the first public release was made available to universities. In 1995, a commercial second release version was made available to the general public. In the late 1990s, Lucent Technologies, who had inherited Bell Labs, dropped commercial interest for the project. In 2000, a non-commercial third release was made under a new open source license. In 2002, a non-commercial fourth release was made under a new free software license. Since then, new release have been made daily that incorporate changes made to the operating system.

The name originates as a pun on the science fiction movie Plan 9 from Outer Space.

If no one objects, I should put it in within the next few hours. Please do not put it in for me as that might be a copyright voilation. ems 12:09, 27 March 2006 (UTC)


 * I like it better, but why risk doing it in for Wikipedia? Why not just ask for permission to use it? Benn Newman 12:15, 27 March 2006 (UTC)


 * The comment about (c) violation is bizzare to say the least, if you post something to a talk page I would guess that the standard license applies at the very least. --Lost Goblin 12:21, 27 March 2006 (UTC)
 * Err, I meant as in I own the copyright on proposal, and as I haven't submitted to the article, it isn't yet license under GFDL. ems 12:25, 27 March 2006 (UTC)
 * Don't be so obnoxious. --Lost Goblin 12:29, 27 March 2006 (UTC)
 * Well unless I actually license it (either by actually licensing it or by submitting it to the article) I own the copyright... thanks to the Congress... Anyway, I hereby place it in the public domain and take no responsiblity for it. ems 12:35, 27 March 2006 (UTC)
 * Why on earth licensing for content writen to talk pages should be any different from content in article pages? Anyway, this subject is pointless and there are a bout a billion things more important things to worry about --Lost Goblin 12:41, 27 March 2006 (UTC)


 * Hmm, whilst I write this comment I see You agree to license your contributions under the GFDL just as I would if I was editing an article. I suggest ems thinks a bit about his/her wikipedia philosophy before contributing here again. -- Jon Dowland 14:43, 3 April 2006 (UTC)


 * Not all the original Unix developers worked on Plan 9, if for no other reason that some were dead (like Ossanna), it was the same team and at the same department though. Also the first sentence about changes in the computing environment is way too POV, use direct quotes from the Plan 9 paper instead if you want to point that out (you are almost quoting it already). The catch phrase isn't very catchy and never cached up very much, so probably we can do without it. There is also redundancy and contradiction between the first and second paragraphs (the first says that Plan 9 was started in the late 80's while the second paragraph says mid 80's. Comment of "operating system of choice" is subjective and POV, it certainly went into production in 89' and was widely used by the early 90's. The comment about the license is confusing, "requirements for free software" can't apply to the OSI guidelines for example. To be honest, I think you could as well start over, it is too messy and confusing. --Lost Goblin 12:20, 27 March 2006 (UTC)

Well, there are some problems with the para. For one thing, it says "in use in 1989", whereas the first internal release listed on the Bell Plan9 website was 1990. You also need to remember that you do not WP:OWN this, or any other, Wikipedia page. -- Gnetwerker 16:03, 27 March 2006 (UTC)
 * You're the one trying to own pages, not us. Sigh... maybe if you did some reading. By 1989 the system had become solid enough that some of us began using it as our exclusive computing environment. This meant bringing along many of the services and applications we had used on UNIX. We used this opportunity to revisit many issues, not just kernel-resident ones, that we felt UNIX addressed badly. Plan 9 has new compilers, languages, libraries, window systems, and many new applications. Many of the old tools were dropped, while those brought along have been polished or rewritten. ems 17:18, 27 March 2006 (UTC)

Submitting. So people fix it up. ems 14:03, 28 March 2006 (UTC)


 * Reverted, you didn't address the issues I listed, if you are lazy don't ask others to do your work for you, already did enough reading what you wrote and spending time pointing out what was wrong with it. --Lost Goblin 14:42, 28 March 2006 (UTC)
 * The only issue I didn't solve was the one of using direct quotes, because I don't see a way of solving that. ems 14:54, 28 March 2006 (UTC)
 * No, the part that you didn't solve is that it is a mess and is not better than what currently is there. --Lost Goblin 15:34, 28 March 2006 (UTC)
 * You never brought that up before. It does need a bit of copyediting, but article already has the cleanup template. And my submit message hinted that it needs copyediting. ems 15:47, 28 March 2006 (UTC)

One of the many problems with this proposal is that it spends half the lead (see WP:LEAD on exposition and synthesis that are not only not sourced and possibly POV, but utterly inappropriate in a leading paragraph. WP:LEAD says: "The lead should briefly summarize the most important points covered in an article in such a way that it could stand on its own as a concise version of the article."  The overall computing environment and in the 1980s and the evolution of Unix are most clearly not a part of a concise version of the article, they are exposition of a particular POV.  The opening sentence as it stands ("Plan 9 from Bell Labs is a distributed operating system, developed at Bell Laboratories by Rob Pike and others, including some of the original UNIX developers") is much, much more conforming to WP:LEAD.  Then the second para goes into a prose version of the release history, when the intro has essentially not told the average reader yet what "Plan 9" even is! By the end of the intro, we are told only that it is (or was intended to be) "Unix [built] out of a lot of little systems" and that it is (was? was intended to be?) "an internal project to start Unix from the ground up with modern ideas" -- as opposed to what, a bunch of old ideas? This is both POV and very poor writing. Really, this proposal is fatally flawed, I cannot see how it could become a lead para here. -- Gnetwerker 22:55, 28 March 2006 (UTC)
 * There is a source: and is in the proposal. It should be direct quotation but I am finding it very hard to keep it short that way. And stating it was developed by Rob Pike and others it totally incorrect, it was developed at Bell Labs' Computer Science Research Center, Pikes involvement was as much as Ken's. Can you please do some copyediting? I am only a en-3 ems 06:12, 29 March 2006 (UTC)
 * I think the motivivation for creating a new operating system when they already had one that worked is important. It also hints at what the o.s. focuses on (in a concise manner). If it is a point of view, it is one that is significant. Benn Newman 16:30, 29 March 2006 (UTC)

I don't care about the Pike reference - though Rob was the one who promoted it the most -- first name on all the papers, gave all the talks (I know, I was there). However, it is of little matter. When I say the proposal is "fatally flawed", that means that no amount of copyediting will fix it. -- Gnetwerker 06:26, 29 March 2006 (UTC)
 * You did in your last edit. Pike's name isn't even on some of the papers, anyway the order of names on papers is pointless. ems 06:54, 29 March 2006 (UTC)
 * Hah! Written by someone who has never been in a "first author" argument! -- Gnetwerker 16:38, 29 March 2006 (UTC)
 * It was written by someone who is tired of your rejecting everything he suggests while not really offering your own alternatives and who probably sees the 'first author' aspect as of mimimal importance. Benn Newman 21:24, 29 March 2006 (UTC)

At last I got around to rewrite the first two paragraphs, adding a few references and updating a bit the following paragraphs. Still lots to do and I still think the whole article is dysfunctional and has to be almost completely rewritten, but until we get there I think this is an improvement, at least I hope it settles the issue of authorship. --Lost Goblin 09:09, 2 April 2006 (UTC)

rfork?
The Impact section mentions that Linux has implemented a "limited form of rfork-like system calls". I think a definition of what rfork is should be included, and how it has been implemented in Linux. (or at least a reference that explains this). I've been a UNIX hacker for 10+ years and have no idea what rfork is.
 * The "Parallel programming" section of the Plan 9 from Bell Labs paper talks about it. Benn Newman 19:44, 30 April 2006 (UTC)
 * Thanks. So does "limited form of rfork-like system calls" refer to Linux's clone (which has flags to choose whether or not to share the file descriptor table, signal handlers, etc)? 209.150.227.67 21:41, 30 April 2006 (UTC)
 * No idea. I'd prefer to not look at--or even be familiar with--Linux' internals. :-p Benn Newman 02:20, 2 May 2006 (UTC)
 * Yes it does. FreeBSD and OpenBSD also have an rfork syscall. Qwertyus 01:42, 3 May 2006 (UTC)
 * Is this somehow related or similar to the Glendix project? If not Glendix should at least be mentioned in the Influenced portion of the See Also section, or perhaps a small paragraph about it somewhere in the article.--Shporang (talk) 17:26, 19 March 2009 (UTC)

Language
Which language is Plan 9 written in? C? C++? —The preceding unsigned comment was added by Frap (talk • contribs) 15:23, 22 March 2007 (UTC).
 * It's written in a dialect of C informally known as "ken C" (in reference to Ken Thompson, who wrote the original Plan 9 C compiler). The most important differences between ken C and ANSI/ISO C, as well as the unusual architecture of the compiler itself, are described in the paper How to Use the Plan 9 C Compiler by Rob Pike. 70.59.201.168 04:32, 19 May 2007 (UTC)
 * An unusual aspect of the Plan 9 C compiler is that a lot more of the work is delegated to the linker than is usual. — DAGwyn (talk) 18:51, 10 December 2007 (UTC)

Missing stuff
The article should have a section on acme and plumbing. Then that should be referred to in countering the criticism about a "simplistic" windowing system. — DAGwyn (talk) 18:51, 10 December 2007 (UTC)

Unicode Support in Plan 9
Although it is announced that Plan 9 has good Unicode support, it in fact only supports the codes in BMP. That is, the codes beyond 0xFFFF may not be properly handled. For example, if we use Acme/Sam open a text file containing codes in SMP, the applications would destroy them. It is a limitation of the system, and could be mentioned somewhere in this article. Thanks.--Realwhz (talk) 16:14, 8 April 2008 (UTC)


 * The two byte Rune size is not a limitation; it is a configuration option. Edit libc.h and recompile. Rune size is also unrelated to UTF, if you want to split hairs. 70.225.171.221 (talk) 07:54, 22 May 2011 (UTC)

download.bat
You can automate daily download while preserving previous version: @echo off del 9atom.iso.bz2.old ren 9atom.iso.bz2 9atom.iso.bz2.old wget ftp://ftp.quanstro.net/other/9atom.iso.bz2 78.90.196.171 (talk) 16:03, 31 March 2011 (UTC).

Production Use?
Regarding the following (from intro) :

"Plan 9 continues to be used and developed in some circles as a research operating system and by hobbyists."

As noted elsewhere in the article Plan 9 has been ported to supercomputers and may be in production use by now. It may no longer be just a research and hobby OS. Anybody know the statu of such projects? — Preceding unsigned comment added by 68.38.196.235 (talk) 15:04, 28 July 2011 (UTC)

Microkernel information in "Background" section
I am planing to remove most of the microkernel related stuff from the "Background" section as it doesn't deserve much more than a mention in passing; and certainly not four whole paragraphs, a paragraph comparing the Plan 9 kernel/design to traditional microkernels might be a good idea, but this is not what the currect monologue about microkernels does, and it should not go into backwrong, which should be more related to the actual backwrond of the Plan 9 devlopment, in what context and why the project was started and so on...

I already removed the paragraph about Mach as it was clearly unecessary. --Lost Goblin


 * Agree. The microkernel is irrelevant here&mdash;a comparison with HURD might be interesting but pointless. In, Ritchie and co. say that 'Plan 9 has a relatively conventional kernel; the system's novelty lies in the pieces outside the kernel and the way they interact'.


 * That said, I wonder if Plan 9 is really dead- it is one of the few OS's ported to Xen after all.

-- Maru Dubshinki 10:21 PM Thursday, 03 March 2005


 * Who said that Plan 9 is dead? I do a replica/pull and get new updates for my systems almost every day, just because Plan 9 devlopers don't like hype and sensationalism doesn't mean that development is not ongoing. --Lost Goblin

Quoting words
There is extreme overuse of "quoting" "random" "words" in this article.


 * Bemasc: There was vandalism by 86.130.210.196, and I've tried to correct it. Alas, I'm a total Wikipedia n00b, and I had a very confusing time of it, with the History page seemingly reporting a different edit sequence every time I looked.  This may have something to do with my caching proxy; I don't know.  In any case, I hope I haven't messed anything up.

Gorram Links not working
Delete all the links that reference to plan9 server at bell labs. Looks like the website got deleted.

Major cleanup
I'm attempting to clean up the article. I've already removed the story-telling elements of the design concepts' descriptions, but a lot of work is still needed. I'm going to make further cleanup here and there, add more references and specifically shorten the external links section. Any feedback and collaboration could be very helpful. &mdash; Dmitrij D. Czarkoff (talk) 10:58, 1 December 2011 (UTC)

The media expelled from External links
I've removed the following entries from External links as they clearly don't belong there:



If anyone have access to these sources, please add them as references to the relevant part of the article or &mdash; if it consumes more time then You can devote &mdash; cite them here, so that I could insert them as references. &mdash; Dmitrij D. Czarkoff (talk) 10:40, 2 December 2011 (UTC)


 * Thank You, AgadaUrbanit, for helping with that one. Now one remains. &mdash; Dmitrij D. Czarkoff (talk) 07:15, 12 December 2011 (UTC)

Plan 9 Wiki
Apparently. Plan 9 has a different definition of "Wiki" than we do: http://plan9.bell-labs.com/wiki/plan9/plan_9_wiki/ --Guy Macon (talk) 11:13, 11 December 2011 (UTC)


 * There are different wiki engines out there. And so what? &mdash; Dmitrij D. Czarkoff (talk) 12:42, 11 December 2011 (UTC)


 * I'm adding the Wiki URL to External links sub-section. AgadaUrbanit (talk) 15:53, 11 December 2011 (UTC)


 * And I'm removing: it is accessible from the official site. Or may be You want to link to every page on the site? &mdash; Dmitrij D. Czarkoff (talk) 15:59, 11 December 2011 (UTC)


 * We also use some internal wiki pages as refs for inline citing. I thought that a direct link to such an external resource would be beneficial for a reader, but on second thought, considering accessibility from the official site a removal sounds good to me. Stay well . AgadaUrbanit (talk) 16:17, 11 December 2011 (UTC)

References format
AgadaUrbanit, You don't really have to format refs; I would do it. &mdash; Dmitrij D. Czarkoff (talk) 14:29, 24 December 2011 (UTC)

References mess
We currently have 36 references with probably more to come soon. Given that some of them are books or journal articles, they take pretty much of space and are not so easy to find unless properly sorted. I sorted them in the order of their appearance in the article, but this practice involves excessive editing, which is not very handy (and ineffective in terms of Wikipedia's storage usage). Thus I would propose to keep references sorted alphabetically according to their names (&lt;ref name="name"&gt;) in case-independent manner. Any comments? &mdash; Dmitrij D. Czarkoff (talk) 14:19, 1 January 2012 (UTC)
 * &mdash; Dmitrij D. Czarkoff (talk) 12:29, 4 January 2012 (UTC)

Brazil version
The Brazil (Operating system) page redirects here, fair enough, but there is nothing on the page about that version. I recall it existed internally to Bell Labs but would like to know how it differed and what differences made it across to the mainstream branch we now have. -- Ralph Corderoy (talk) 18:35, 5 August 2012 (UTC)


 * According to this, Brazil was a nickname for Plan 9 release 3, an according to this, "the current version of Plan 9 was called "Brazil" for a while". --Guy Macon (talk) 19:15, 5 August 2012 (UTC)

No information on kernel architecture type
Microkernel? Monolithic (possibly with modules)? Hybrid? Other? 76.10.128.192 (talk) 11:14, 29 June 2012 (UTC)


 * 9fans usually either consider it to be a monolithic kernel or don't find the distinction useful. 50.122.38.241 (talk) 12:03, 25 August 2012 (UTC)

Unicode ⇔ UTF-8
Recently the article suffered from a minor edit war about Unicode support in Plan 9. As long as the trimmed down "UTF-8" version doesn't allow one to judge on problems the editor sees with previous version, I would like to start a discussion here, so that we could find out the core of a problem and address it the way that would benefit everybody (and topic coverage in the first place). — Dmitrij D. Czarkoff (talk•track) 19:35, 6 October 2012 (UTC)

Original Bell Labs MIT-like or Lucent Public License?
This page from Plan 9 says "license: original Bell Labs MIT-like or Lucent Public License. Now according to this non-reliable source, dual licensing means that the source code can be licensed under one of two separate licenses, not that it's simultaneously licensed under both, and our article on Multi-licensing seems to agree. So, should we say that the Plan 9 license is GPL-compatible?

To further confuse the issue, Richard Stallman says that the current license of Plan 9 does qualify as free software (and also as open source). (Written in 2000, last updated 28 February 2013) Doesn't RS equate "free" with GPL-compatibility? --Guy Macon (talk) 08:46, 20 April 2013 (UTC)


 * IANAL, but imho all this GNU vs. BSD vs. MIT license discussions probably do not belong to this article, which should be about Plan9. I'm going to remove "however not a GNU" clarification, since this article does not claim that Plan9 is covered by GNU. AgadaUrbanit (talk) 16:06, 20 April 2013 (UTC)


 * Also IANAL. We should not say that Plan 9 is GPL-compatible, and "free" is not the same as "GPL-compatible". Long ago, software could be, and was, licensed to different licensees using different licenses, depending on several factors. (It still can be.) I worked in a lab that had three versions of Unix, running on different machines, each with a different license. htom (talk) 18:40, 20 April 2013 (UTC)


 * Good call on the removal. The part that was removed really doesn't add to the article.


 * That being said, the article still says "The full source code is freely available under Lucent Public License", while at least some parts of the the Plan 9 documentation say "License: original Bell Labs MIT-like or Lucent Public License. That seems to imply that we should say "The full source code is freely available under a MIT-like License or the Lucent Public License". On the other hand, this bell labs page only mentions the Lucent license. Could it be that only the Unix port of Plan 9 is under the MIT license?


 * BTW, the "MIT-like" language links to this swtch.com page, which is clearly this Princeton University variant with the Princeton-specific part removed. According to that page, all of the MIT variants are functionally identical. --Guy Macon (talk) 19:24, 20 April 2013 (UTC)


 * The MIT-like license is for Plan9port, not Plan 9 itself. Plan 9 is under Lucent Public License 1.02. Diyoev (talk) 01:49, 21 April 2013 (UTC)

Principles of Plan 9
There has been a recent hatchet job on this article.

In particular, the removal of the key principles (sourced) behind the architecture of Plan 9. I can only assume that this is due to youthful ignorance of, not Plan 9 (how many people did ever use it hands-on?), but of the state of other contemporary systems. Needless to say, they're now edit-warring to repeat this deletion, because their deletion of the logo was opposed. I see 9front has gone as well.

It used to be a Good Article too 8-(   Andy Dingley (talk) 19:03, 13 February 2015 (UTC)


 * First off, I would like to remind you of WP:AGF. Calling edits that attempt to clarify an article's contents a "hatchet job" is in blatant violation of that principle.
 * Second, let me explain my removal of the "principles" of Plan 9.

"Plan 9 is an evolution of UNIX design concepts:
 * 1) all objects are either files or file systems
 * 2) communication is over a network
 * 3) private namespaces allow their owners to access local and remote processes transparently"


 * From the very first sentence, this is unclear and unworthy of a Good Article. Are these Unix design principles? No. The first one is, to some extent, the rest are not: in Unix there is one namespace per machine, and networks were added as an afterthought (multiple times with different APIs, never integrated with the FS). If we're going to compare to Unix, then let's describe what Unix does, and how Plan 9 generalizes that. The second principle, "communication is over a network" is a slogan that is not explained. What does that mean, that processes cannot talk to each other unless their messages pass through a network?
 * As everyone can see in the diff, I replaced this with some (well-sourced) prose and examples that actually explains principles 1 and 3 while ignoring the strange principle 2 that makes no sense without an explanation. (Maybe the source actually elaborates on this, but I've no access to it.) Q VVERTYVS (hm?) 20:14, 13 February 2015 (UTC)


 * "Are these Unix design principles? "
 * And that is why you should leave this article well alone.
 * Plan 9 is not Unix. It is something different. If you don't even appreciate that much, no wonder you've made such a wreck of it. The "Delete things I don't understand" approach so rarely works out well.
 * Plan 9 began from these principles (with a load of pre-existing Unix background). It was an axiomatic decision (the merit of which has been argued since) that everything would be presented as a file / file system. It was also the first OS (in a still broadly pre-Internet world) to assume that the Internet was/would become ubiquitous and that networked devices (even if individually trivial) needed to operate in that context first and foremost. Plan 9 was misunderstood for a decade or so and frequently criticised as being a poor desktop PC OS - completely missing its point. The clever aspect was to recognise (at this initial axiomatic level) that this would also require the third principle, of local scoping rules. All three of these were novel for Plan 9: even with contemporary OS that had "good" networking, these were always as an alternative (and with an alternative API) to the "fundamental" means of working via local device access. This was novel for Plan 9: it was not part of Unix, nor even Windows, MacOS or even VMS. Andy Dingley (talk) 20:43, 13 February 2015 (UTC)


 * You're missing my point entirely, and it would serve you well to cool down before telling me what to do.
 * My point is that the old text can be read as if these three points are Unix design principles. You don't need to convince me that they're not. You should be worried about the reader who is not going to understand why any of this is important.
 * Read the old text again, and try to take the position of someone who does not know what Plan 9 is. Can you tell what the list I removed is trying to convey? Is it listing Plan 9 principles only, or principles shared with Unix? "Communication is over a network"? What does that mean? Why wasn't Rob Pike's explanation of the two(!) design ideas of Plan 9 cited instead, until I added it a few minutes ago? Q VVERTYVS (hm?) 21:14, 13 February 2015 (UTC)
 * Have you not heard of copyediting? Andy Dingley (talk) 21:20, 13 February 2015 (UTC)
 * Incidentally, for those previously unfamiliar with Plan 9, a clearer explanation is in ESR's paper (later part of The Art Of Unix Programming) 'Plan 9: The Way the Future Was'. This is written as an explanation for those unfamiliar with it, not like the Bell Labs papers that were generally written for those who'd been following its history all along. Andy Dingley (talk) 21:31, 13 February 2015 (UTC)


 * I have not liked the past presentation of the 'Design concept', as well as other parts, in this article. The reader was thrusted against three bullet points that had no tangible contexts meaningful to a general audience. E.g., "all objects are either files or filesystems" is devoid of meaning and in such simple form, actually false.  The new prose states these same highlights in better fashion in relation to known features of the predecessor system Unix, from which they evolved, in this case from the notion of a /dev filesystem with device nodes representing peripherals. Already in Unix there was the notion of this being extended to the network, in RFS-based networking (only in SVR3 and 4). Providing the context is important which the old version did a much poorer job of.  So, sometimes a hatched job is really needed for improvement. Kbrose (talk) 21:51, 13 February 2015 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 1 one external link on Plan 9 from Bell Labs. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Corrected formatting/usage for http://www.techradar.com/news/software/operating-systems/raspberry-pi-operating-systems-5-reviewed-and-rated-1147941

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.—cyberbot II  Talk to my owner :Online 15:05, 1 April 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 external links on Plan 9 from Bell Labs. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20080706034735/http://9fans.net/archive/1995/07/16 to http://9fans.net/archive/1995/07/16
 * Corrected formatting/usage for http://soc.dev.gentoo.org/~rps/plan9/
 * Added archive https://archive.is/20130221074800/http://9fans.net/ to http://9fans.net/

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

Cheers.— InternetArchiveBot  (Report bug) 16:24, 26 July 2017 (UTC)

Plankalkül
The only other programming language I know of that starts with 'Plan' is Plankalkül, Konrad Zuse's ancient (1948) programming language. There's no allusion to that? Mcswell (talk) 02:38, 3 May 2019 (UTC)

9p.io
"9p.io" is still in Alcatel-Lucent trade dress, and does not appear to be associated with Nokia Bell Labs. Would it be better (safer, more accurate) to link to the Wayback Machine rather than an unverified mirror site?--NapoliRoma (talk) 22:41, 24 July 2019 (UTC)

citation needed on creation date
The article states "Plan 9 from Bell Labs was originally developed, starting in the mid-1980s," (I've added [citation needed] to this). The citation at the end of the sentence covers the rest of the sentence, but does not mention when development started. As far as I can tell, "mid-1980s" claim in the article originated uncited in this April 2006 edit (which was just weaseling out of an older uncited, but more falsifiable, "1989" claim). Most of the authoritative sources I've found state that it was started in the "late 1980s" (for instance: https://9p.io/plan9/about.html). The best I've found for a concrete year is that https://doc.cat-v.org/plan_9/1st_edition/designing_plan_9 states that they worked on it in 1987 and implies, but does not explicitly state, that there was no work before that. ~ LukeShu (talk) 05:37, 28 December 2019 (UTC)
 * Their own history document is clear and agrees with the others you found. I've updated it to "late 1980s". Elizium23 (talk) 05:41, 28 December 2019 (UTC)

"Open Source"
I think its important to use FOSS or Free software rather than Open source, which is an expression generally used by corporate projects. Now plan9 is a community project so the FOSS term just applies there. --Comrade-yutyo (talk) 09:17, 1 April 2021 (UTC)
 * Most of the community-developed operating systems (e.g. Debian, OpenBSD, Haiku (operating system), ReactOS, FreeDOS, OpenIndiana and many others) also use "open source". I think it's a good idea to be consistent across the articles of different operating systems. Vt320 (talk) 13:50, 1 April 2021 (UTC)
 * Source model is "open source", but I think it's important to add in the lede that the software is free and open source (as recommended by FOSS task force). I made an appropriate change. — K4rolB (talk) 10:12, 17 July 2021 (UTC)

List of distributions?
Having a list of current distributions of plan9 seems like a good idea, considering that's where all the current work is being done. Binarycat64 (talk) 03:39, 9 November 2021 (UTC)

Infobox
Thank you to whoever did the “License” section in the Infobox that clearly delineates the evolution of Plan 9’s licensing by year; well done! —PowerPCG5 (talk) 11:02, 7 January 2022 (UTC)