Wikipedia talk:Notability (software)

From Wikipedia, the free encyclopedia

Centralized discussion
edit • talk • log • watch
Discussions
Broad RFC on the nature and purpose of notability and the GNG Discussion on the current notability guidelines Proposal to add 'class C' to the article grading system Revisiting status of WP:Attribution Centralized discussion on terrorism related articles ArbCom proposal to create a Sourcing Adjudication Board Discussion to work out the implementation details of flagged revisions Discussion over the approval of bots Wikipedia:Plot summaries Global blocking on Wikimedia projects SSP/RFCU merger proposal Wikipedia:Notability (criminal acts)
Archive

Archives

• Wikipedia talk:Notability (software)/Archive01

[edit] Archived

I just archived this talk, I'll try to get a summary together. - brenneman 23:48, 15 January 2007 (UTC)

[edit] Distributions

The thing that led me here was the "being bundled = notability" criterion, by way of Wikipedia:Articles for deletion/Group-Office (third nomination) so I'll start by summarising those arguments. I'm going to copy/paste the relevent section here and then distill them to the juicy nuggets. - brenneman 23:57, 15 January 2007 (UTC)

[edit] Linux distributions

A software package should get bonus points for being in an OS distribution (Linux, FreeBSD, etc.). For the more selective distributions such as Fedora Core or Ubuntu main, just being in the distro should make it notable. Debian is too big for a software package to be included in Wikipedia just because it has a package in Debian; the Debian Popularity Contest is a great resource; we could choose a percentile above which packages are considered notable. —Quarl ^(talk) 2006-01-22 19:37Z

Agree. Inclusion in a major mainstream distro is notable. - Just zis  Guy, you know? ^[T]/_[C] AfD? 22:36, 25 January 2006 (UTC)

How is the list of Linux distributions determined? The use of the word major appears to make that criteria subjective. DistroWatch lists the top 10 [1] as:

1. Ubuntu
2. Mandriva Linux
3. SUSE LINUX
4. Fedora Core
5. Debian GNU/Linux
6. Knoppix
7. MEPIS Linux
8. Gentoo Linux
9. Slackware Linux
10. Xandros Desktop

Also, the criteria is qualified by "The more packages a distribution includes, the less notability is implied by inclusion in that distribution." It would be useful to provide package count stats or links to stats to help wikipedians see which distro's hold more weight. John Vandenberg 20:19, 15 January 2007 (UTC)

Just added that list to a new section on linux distribution. I was looking for it earlier there. So thanks :) FT2 ^{(Talk | email)} 03:59, 16 January 2007 (UTC)

[edit] An alternative view

I strongly disagree with the idea that inclusion in any distro constitutes notability. I'll illustrate why with some examples from what's described above as one of the more selective distros, Ubuntu - do any of the following warrant inclusion in an encyclopedia?

• cookietool - A team of programs to help you maintain your fortune database
• defendguin - Defender clone with penguins
• dvorak7min - Typing tutor for dvorak keyboards
• tablix - high school timetable generator
• kazehakase - gecko based web browser using GTK

Now, to me at least, that seems ridiculous, but my interpretation of the comments above is that they're saying these packages are automatically notable and worthy of an article on Wikipedia. Where am I going wrong, and what's wrong with the idea that something is notable if it attracts enough attention for independant and reliable parties choose to write about it? CiaranG 13:29, 16 January 2007 (UTC)

Those definitely are obscure, and I presume are NN. However, you have only looked at the Ubuntu distro (which isnt mentioned in the guideline), and at least the first of those can be discounted by strict interpretation of "and the maintainer of the distribution is independent from the software developer", because the packaging appears to be maintained by one of the cookietool software developers. According to the google cache version of http://aminet.net/util/misc/cookietool.readme, cookietool is developed by "Wilhelm Noeker & Miroslav L. Baran" and Miros/law L. Baran is also the major maintainer of the debian package, which makes its way into ubuntu. So that package is only in debian/ubuntu because Miros/law L. Baran puts in the effort required to keep it there, rather than it being considered necessary by the distribution management group. John Vandenberg 14:33, 16 January 2007 (UTC)

Fair comment - I chose Ubuntu in response to the comments above, rather than the existing guideline. Perhaps cookietool was a bad example. How about defendguin, which is also part of the Debian distro, and seems to satisfy the "maintainer of the distribution is independent from the software developer" clause. I can't see for a minute how "defendguin 0.0.11" is notable, yet because of this 'inclusion in a distro implies notability' concept, the guideline says it is, and that's my point - what's wrong with the idea that something is notable if it attracts enough attention for independent and reliable parties choose to write about it? What's the purpose of this 'bundling cause' at all? (I have taken the trouble to read the archives first, honest, but my question is not answered there) CiaranG 15:06, 16 January 2007 (UTC)

defendguin is developed by New Breed Software (which is somehow related to Bills Games ([3]) and they have a partial list of platforms that defendguin is available for here. It has a http://sf.net/projects/defendguin/, but that appears out of use. Anway, here is a list of platforms that I have compiled from a google search, attempting to verify the people involved:

• Microsoft Windows email from bill to SDL mailing list, answered by Matt, who is now listed as the porter on the games download list
• Mac OS X [4]
• Debian (packager Christian T. Steigies)
  • Linspire [5] (maintained by debian)
• AltLinux (packaged by Victor Forsyuk)
• Mandriva (last update 2002)
• Klik (packaging) (pretty screenshot, "This recipe has no maintainer" ... apparently doesnt need one as recent comments indicate all is well)
• Gentoo (various packagers)
• SLAX ([6])
• NetBSD (packaged by Jeremy C. Reed)
• FreeBSD (various maintainers)
• OpenBSD (maintained by The OpenBSD ports mailing-list <ports@openbsd.org>)
• GnuWin II ([7])
• BeOS ([8])
• QNX (a contrib package)
• PlayStation Portable [9] [10]
  • another PSP port [11]
• GP32 ([12])
• and others (e.g. DreamCast)

It has also been the subject of reviews

• http://linux.about.com/cs/linux101/g/defendguin.htm
• http://www.marcelgagne.com/cwl102004.html
  • published in Linux Journal: http://www.linuxjournal.com/article/7663
• http://www.happypenguin.org/show?Defendguin (takes a while to load; huge images by the looks of things)

It is listed on List of portable computer games. I dont think this game needs its own article, but it seems to pass WP:V and WP:N (see http://www.newbreedsoftware.com/defendguin/reviews/). All that remains is how will WP:SOFTWARE treat it. In my opinion, a game "such as this" should be covered by either New Breed Software or Bills Games as there are quite a few other games in that catalogue that are of the same ilk. Also it could be mentioned on Defender (arcade game), in the same vein as Winmine. John Vandenberg 20:31, 16 January 2007 (UTC)

Thus this doesn't need the bundeleing clause, that's what you're saying right? - brenneman 21:52, 16 January 2007 (UTC)

That's my interpretation as well, which makes me re-iterate my initial point - what is that clause for? Either the software's notable, or it's not. I do consider myself severely rebuked for suggesting Defendguin is non-notable though. You live and learn. :) CiaranG 21:57, 16 January 2007 (UTC)

The `bundling' clause? At this stage, I'm not ready to say that the clause isnt useful. If anything, I think that defendguin proves we need something more than WP:N because currently defendguin is notable enough that it its inclusion as a separate article could stand irrespective of the bundling clause (my guess is that an Afd on a well written defendguin article would be contentious -- I'd vote delete/merge but the guidelines would not be on my side).

At this stage, I am pretty confident that we can define inclusion in a notable & selective distro or rankings so that it provides a precise criteria for inclusion -- doing so allows people to know when they are on safe ground. For example, we might list a few groupings such as Gentoo Reference Platform as sufficient for inclusion. For Redhat/Fedora, the inclusive list should be packages that are available in the default package manager (the one that lists only major apps in a few groups -- i'll find the name if nobody else can help me out). I'm happy do the leg work required to find groupings that can be inspected by yourself and others to see if they are suitable.

I would also like to see some way to exclude defendguin, and my crystal ball tells me that we can word this clause so that omission from a notable & selective distro or rankings also clearly defines when software is not notable. This will be more difficult but also more useful in Afd's. FT2's first two proposed criteria below are a good start, but leave out how to define "significant". If we can find a few reputable software lists that cover 75% of the exist articles already in Wikipedia, we could probably add a strict rule to WP:SOFT that requires that software must have been recommended by a notable WP:CORP in a review that evaluates many similar packages at once. That type of wording automatically allows articles that only cover a single piece of software to be discounted as a claim of verifiability/notability. Then we can Afd the other 25% :-)

One major problem with this clause is web applications are rarely distributed via package management systems. One way to deal with that is to require that web applications must be hosted by a notable Application service provider (WP:CORP) or prominently used by a notable website (WP:WEB).

I'm definitely not comfortable having this clause removed out of hand; they will need to present a well researched case(s) where the bundling clause is problematic, and better wording doesn't solve the problem. John Vandenberg 09:23, 17 January 2007 (UTC)

Concerning web applications, I fail to see why they should be required to be hosted by a notable ASP when the bundling clause (thought to be applicable only to installation-based applications) is only presented as an additional criteria of notability (ie not mandatory). Most web applications exist on its own without reliance on another ASP. However, like the bundling cause's effect to installation-based applications, an additional criteria that relies on being hosted by a notable ASP (such as Salesforce.com) would be appropriate for web apps. --Pkchan 03:32, 18 January 2007 (UTC)

[edit] Notability

The question is, under what situations would (stand-alone, integrated or systems) software be notable. I can think of a few. Can we look closer at them? These are my suggestions for notability criteria:

1. They are included as significant ?separable functional modules or components of a notable operating system, or (for those systems distributed linux-style) as significant ?separable functional modules or components of a distro. (Captures component parts of systems, such as DirectX, iptables,samba, but also key interfaces such as interprocess piping)
2. They are separate software bundled or distributed as standard within a major OS or distro (captures programs like gimp, or windows media player)
3. They are leaders or notable in their specific software niche, which itself is not so narrow as to exclude notability (for example trillian in IM packages, Opera in browsers, AMIBIOS in PC bioses, PalmOS in handhelds, open office and Microsoft Office in office productivity packages)

   • Note: minor niches may require their leading or well known software to be summarized in the main article as separate articles may not be necessary or appropriate. (prevents proliferation of minor stuff)

4. They are notable for being key, major, and well known popular add-ons, plug-ins or modules for notable software. Most software of this kind should be merged into the main article; however in a few cases there may be a basis for a separate article for the add-on, subject to editorial discussion. (handling of notable add-ons, usually within the main article, occasionally may deserve own article)
5. They are or were historically relevant or of enduring significance in the history of software. (captures packages which, though not relevant as software, are relevant for having broken new ground, introduced new technology, or new concepts etc)
6. An operating system (whether kernel, distro, or packaged windows style 'all-in-one') is considered notable if it is major and established or of significant popularity within the market as a whole or a specialist niche, and has obtained a measure of renown and acknowledgement as being notable (beyond mere reviews and awards of approval) amongst credible commentators of the industry and market as a whole, or within its niche. (Trying to capture criteria for OS notability. Broadly, any OS, of whatever style, must be notable within the industry and/or its niche. It must also be acknowledged as such by credible commentators, which doesn't just mean "magazine X did a review" or "website Y gave it an award". If its notable it will probably have people saying specifically, that it is notable, in some way or other, not just that it's popular or "we did a review".)

Exceptional circumstances only:

1. They are notable in the eyes of non-owners and non-users. This is rare, since most software of notability will fall under one of the above headings.
2. Popularity alone, unless extreme, will usually not be a measure of notability, since any software popular enough to be notable would be likely to be notable under one of the above.
3. They have a notable role or function, such as Narus or ECHELON, which could be considered notable even if it wasn't a key player in its field, due to the situation or context it was used in. (notable by exceptional context - ie circumstances or usage)
4. Other reasons for notability in exceptional circumstances may be considered but will need exceptional justification.

How do people like that as a starting point of a framework for software notability? FT2 ^{(Talk | email)} 04:36, 16 January 2007 (UTC)

Number 2 (bundled or distributed as standard within a major OS or distro) would mean we will accept a separate article for any obscure program in MS Windows; I see that Notepad, Winmine, Regedit, and Microsoft Calculator already have articles, but there are many more that are listed in Programs => Accessories that we dont have articles for (yet). e.g. I think Task Scheduler should be either an article or a redirect, but I dont think the XP Files and Settings Transfer Wizard deserves an article. John Vandenberg 14:49, 16 January 2007 (UTC)

And otherwise, in general as a starting point for discussion?

You do raise a good point: what criteria could we apply that would include Task Manager but exclude (for example) the files and settings transfer wizard you mention? That'd be an issue in any notability criteria, so maybe its a good one to ask ourselves, how we'd draw a line to make one pass, the other not (if thats whats wanted)? FT2 ^{(Talk | email)} 13:33, 17 January 2007 (UTC)

Number 3 (They are leaders or notable in their specific software niche, which itself is not so narrow as to exclude notability) - they're notable if they're notable? Can you define notability by using notability as one of the criteria? Similar issues with several of the later points, and in general, my personal view is that it's far all too subjective. CiaranG 13:53, 17 January 2007 (UTC)

Ultimately, most of the notability criteria in this field are likely to have a strong element of subjectivity. Notability criteria are used as a way to judge individual cases, with grey shades in some areas and decisive rules in others. I can't see any criterion that will distinguish explicitly a line of notable/non-notable for all software. But by giving an outline of what sort of thing we're looking for, we will have enough to judge individual AFD cases, and in practical terms, that's really all that this guideline will be required to be applied to.

I'd say that number 3 is a usable and sensible criterion. If you want to have an article on Opera, on the basis it's a notable browser, then it needs to be notable amongst browsers, and also, browsers as a type of software need to have a degree of notability as a type of software. If browsers are a very minor kind of software then even their market leader may not justify its own article, and should just be described in the article on browsers instead. Thats what #3 is saying. Seems pretty commonsense and logical.

I think if we give an outline that covers and gives an overview of the main themes we're looking for, we can trust editors judgement in deletion debates and the sources they cite to support fitting into those criteria (or not fitting), for the rest. Wikipedia has always had a preference for minimalizing detailed instruction creep. FT2 ^{(Talk | email)} 21:11, 17 January 2007 (UTC)

[edit] Proposed merger

Radiant! placed the merger tag on both this article and Wikipedia:Game guide.

• Strong oppose They are not even remotely the same thing. And should NOT be merged. Havok (T/C/e/c) 12:34, 17 January 2007 (UTC)
• First of all, this is not a vote. Second, most of "gaming guide" refers to computer games, which are software. Third, notability is not merely about deletion-or-not, but also about organization of material; note how WP:FICT talks about plot summaries and such. And fourth, the two proposals combined have a much better change of getting consensual support. >Radiant< 12:42, 17 January 2007 (UTC)
  • We are discussing if it should be merged or not, and I oppose it. If you proposed the merger on grounds that much of the content in WP:SOFTWARE was the same as in Wikipedia:Game guide then I would understand it, but as it stands now, they are two totally different things. The reason Game guide was created to begin with, was to be more specific as to what would be considered breaking WP:NOT, as WP:NOT isn't clear enough on this. Neither is WP:SOFTWARE. Havok (T/C/e/c) 13:38, 17 January 2007 (UTC)
    • Indeed, this page isn't clear enough on that. But it should be. >Radiant< 13:41, 17 January 2007 (UTC)
      • But as noted below, the problem is also that Software refers to more then just computer games. Which might dilute the whole point of having the game guide rules, which are there to protect articles from deletion by setting clear examples of what is allowed and not in gaming articles. Havok (T/C/e/c) 13:46, 17 January 2007 (UTC)
• Comment The concern is that very different specific criteria will probably apply to these. For example, distribution with a major OS will not be a factor for games, and recognition on a major gaming site or top 100 sales on a game chart will not be a criterion for most general software. I don't have a problem with them being in one guideline, as Radiant suggests, to prevent proliferation of many minor notability criteria, but if they are then they need to be two different sections, "general software" and "computer game software" (and possibly other sections for other types of software). That said, I also don't have a problem with them being separate guidelines, due to the likely near-complete lack of overlap of criteria in these two specialisms. FT2 ^{(Talk | email)} 13:31, 17 January 2007 (UTC)
• Strong oppose - I believe that the criteria for a video game is much different that general software, even if a video game is a form of software. Greeves ^{(talk • contribs)} 04:42, 27 January 2007 (UTC)
• I just noticed that it has been five days and I believe that we have consensus. Without any other supporting Wikipedians, I believe that this proposed merger is closed. Greeves ^{(talk • contribs)} 04:45, 27 January 2007 (UTC)

[edit] Test case: Alcohol 120%

If you all want a test case, take a look at Wikipedia:Articles for deletion/Alcohol 120%. One user argues that (1) WP:SOFTWARE applies and (2) on-line reviews and reviews on download sites qualify Alcohol 120% as notable.[13]

I am not really familiar enough with your proposal to judge, but am curious how you see it playing out. Thanks, TheronJ 19:39, 23 January 2007 (UTC)

[edit] Games

How would the notability of a game be decided then? Shrumster 21:08, 23 January 2007 (UTC)

[edit] Make it clear

This criterion includes published works in all forms, such as newspaper articles, books, user guides, television documentaries, and full-length magazine reviews 2 except for the following:

• Media reprints of press releases, other publications where the author or manufacturer talks about the software, and advertising for the software. Newspaper stories that do not credit a reporter or a news service and simply present company news in an uncritical or positive way may be treated as press releases unless there is evidence to the contrary. 1

Make it clear by, for example saying, any third party newspaper/magazine/tele reports instead of such long, detailed but ambiguous words. --Deryck C. 14:13, 1 February 2007 (UTC)

[edit] distro

Based upon the comments on this page, I'm going to remove the distribution section. Anyone going to scream? - brenneman 03:27, 10 February 2007 (UTC)

I'm certainly in favour - I think it's unnecessary and clouds the issue. CiaranG 07:31, 10 February 2007 (UTC)

Which section of discussion on this page are you referring to? I can count two editors in favour of removal but one editor against removal out of hands in the section #An alternative view and I don't think that can be seen as consensus. --Pkchan 09:05, 10 February 2007 (UTC)

It's also worth looking at the sections Core products and operating system distributions and #insanely biased towards free software. for more apparently in favour of removal. There are also a number in other sections who clearly would not be in favour of removal, so I do agree further specific discussion here is warranted. CiaranG 09:12, 10 February 2007 (UTC)

[edit] Rejecting this proposal

(1) There is little in this proposed guideline which is not already covered in WP:ORG under products or genereally at WP:N

(2) This has become an example of unneeded [[WP:CREEP| instrucution creep

(3) After sufficient discussion no consensus has been reached.

Therefore, this should be tagged as rejected under Wikipedia:Policies and guidelines which state: "A rejected page is any proposal for which consensus support is not present, regardless of whether there's active discussion or not. Consensus need not be fully opposed; if consensus is neutral on the issue and unlikely to improve, the proposal is likewise rejected. Making small changes will not change this fact, nor will repetitive arguments. Generally it is wiser to rewrite a rejected proposal from scratch and start in a different direction."

--Kevin Murray 03:37, 1 March 2007 (UTC)

• That's overdoing it a bit. The text here is reasonable and seems to match the way AFD works. It would probably be better off merged somewhere, because of the overlaps, but having some kind of guideline for software is a useful idea given the prevalence of software on the internet. >Radiant< 09:22, 1 March 2007 (UTC)
  • Radiant, there isn't really anything here except a restatement of the primary criterion and some special cases which are statements of the obvious and woule surely meet the prime criterion. This is meaningless creep going nowhere. --Kevin Murray 23:21, 15 March 2007 (UTC)
    • Except that not everybody likes the "primary criterion" and we do have succesful pages for WP:WEB, WP:CORP, et cetera. There's no need to preemptively reject this based on the assumption that the whole lot may become merged at some point. >Radiant< 09:32, 16 March 2007 (UTC)

There seems to be no progress toward consensus here since the end of January. This seems to be dead and should be tagged as rejected. --Kevin Murray 16:46, 26 March 2007 (UTC)

[edit] Stacked against F/OSS Software (with case in point)

It's biased *against* F/OSS software - OSS doesn't get reviewed in magazines for example, even if it has a few million users, you have to be up to the point of Debian or Spamassasin to be 'notable' acording to this guideline.

Case in point, somebody has tagged the E107 (software) with both this guideline and the {{advert}} one, so now there's a problem, notability could *easily* be established with a link to http://www.packtpub.com/article/open-source-content-management-system-award-winner-announced for example - but now it looks even *more* like an advert. Even though it's OSS - advert, funny. So no now with this perspective in mind there's two guidelines that are both strongly add odds with each other and frankly, unworkable.

Don't get me wrong, the page is nasty and needs fixing, but this guidline just makes it that much more akward --Streaky 14:05, 12 March 2007 (UTC)

[edit] Don't reject this

There seems to be a trend whereby if a proposed notability guideline has not been accepted as a guideline, an editor will tag it as rejected. I object to this trend, and I would rather see this guideline accepted than rejected. --Metropolitan90 14:31, 16 March 2007 (UTC)

• It's not really a trend, it's a single editor doing that :) at any rate, "software" is a worthwhile category to have some guidance on. Perhaps you could cite some AFD debates that show how it works out in practice? >Radiant< 10:16, 19 March 2007 (UTC)

Per the WP Policy: "A rejected page is any proposal for which consensus support is not present, regardless of whether there's active discussion or not. Consensus need not be fully opposed; if consensus is neutral on the issue and unlikely to improve, the proposal is likewise rejected. Making small changes will not change this fact, nor will repetitive arguments. Generally it is wiser to rewrite a rejected proposal from scratch and start in a different direction. --Kevin Murray 16:52, 26 March 2007 (UTC)

• Yeah yeah yeah, you don't have to keep citing that to me, I know what it says since I wrote that page. At any rate, your comment that there's "no progress here since the end of January" seems contradicted by the fact that this talk page was edited last week. It seems this could be a worthwhile guideline if people would be interested in it (which would require some advertising) so {{historical}} seems more appropriate. >Radiant< 16:58, 26 March 2007 (UTC)
  • Radiant, I didn't say that there was no discussion; I said that there was no progress. If you wrote the quote, you should remember that it says "even if there is active discussion." Plus you should be flattered at my love of your fine word-crafting. Cheers! --Kevin Murray 03:30, 27 March 2007 (UTC)

[edit] We need this page

Is there any way to whip this page into shape? I really think it's an important guideline, and should become official. Andre (talk) 07:44, 5 July 2007 (UTC)

One approach would be to discuss a substantial number of actual examples. Then we could deduce what notable software has in common and how these properties can be measured. My impression has been that so far, we've been writing the rules first only to be surprised at the results when they were applied. Rl 08:22, 5 July 2007 (UTC)