Talk:Comparison of IRC clients/Archive 2

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

Platform column

The platform column is pretty useless at the moment. It just lists the processor architecture not the actual platform. "Platform" means a combination of processor architecture, underlying OS, and vendor. Things like Android, Solaris, AmigaOS, Ubuntu etc. need to be added.

If you listed *that* information, the web page might be useful to me. I came here to find an IRC client for a foreign platform that I didn't know anything about (something called Windows). Instead I ended up downloading Opera as I know that it has a built-in IRC client. 87.194.208.119 (talk) 10:36, 14 January 2010 (UTC)

he? solaris is in, amigaos also, android is a Linux, ubuntu also: so why not looking in these columns? these table/comparison is more cluttered that every other comparison i saw in wikipedia! mabdul 23:12, 14 January 2010 (UTC)

Somebody please consider removing the platform column: A quick search on Debian's Package site will reveal most clients are for all the supported Platforms. The chart about OS support is much more useful in this case. --99.41.104.240 Thu Jan 20 00:09:13 UTC 2011 —Preceding unsigned comment added by 99.41.104.240 (talk) 00:09, 20 January 2011 (UTC)

Orion (IRC client)

Notability not established, and inappropriately redirected to this article. I've found no reliable, independent, notable sources, discussing this IRC client in news, magazines, journals, or books. The blue link to [[Orion {IRC client)]] is a redirect to this very article, which cites no sources about this client. The WP:GNG General Notability Guideline, yes, applies to articles, but the Orion IRC Client lacks sufficient WP:RS to possess its own article. For comparison, see the rather well-sourced Chatzilla article which cites, yes, the developer and Mozilla sites, but also includes magazine articles, reviews, and books. Please help by locating and listing WP:RS which can be used to create an article for Orion, and any other non-article-yet-possibly-notable clients. IMHO, "notable enough for this list" should mean "has its own article in which notability is established." Otherwise, this comparison's entries will always be vulnerable to summary deletion, one by one, by deletionists, or even quite reasonable "no source? delete" editors. --Lexein (talk) 18:49, 30 January 2011 (UTC)

http://tellini.info/software/orion/ I can't figure out how to add that correctly to references. Also, can you please fix the formatting/links on the first table's Orion entry? I was trying to make the tables consistent.

71.214.52.97 (talk) 14:23, 6 February 2011 (UTC)

Bah nevermind - just read the *sshattery above. I guess they'll do a bulk fix, anyways when the article is repaired. Wow, mass deletes of sourced materials FTL!71.214.52.97 (talk) 14:36, 6 February 2011 (UTC)

Thanks. However, Tellini is the author, a primary source not a reliable (nonblog, nonforum, nonwiki) independent (completely unrelated to Tellini), notable (established, published) 3rd-party source. The Orion client has not been reviewed or even mentioned in any magazines or related blog, journals or related blog, newsletters, or books. One such review will be good, two would be better; then I'd call it sourced. At the moment, it should be removed from all this article's tables, IMHO. All new software suffers from this sourcing and notability battle. At the moment, I'm not bothered that Orion doesn't have an article - it's definitely not WP:GNG generally notable enough for that - I'm just looking for any kind of WP:RS. I hope you understand that I have the same problem with any entry in this list which lacks an article and any WP:RS. --Lexein (talk) 02:22, 7 February 2011 (UTC)

Good grief... Why is this argument still turning up here again? This stuff has been discussed to death above. The notability guideline does not limit what can or cannot be written about in an article. The WP:NNC section of the notability guideline is quite explicit on this and trying to limit article content based with the WP:GNG was not a view supported by one of the recent RFCs.

As long as we have verifiability there is nothing preventing us from discussing a particular software program in relation to others. For basic features and functionality of a software program, a primary source is sufficient. As long as we don't attempt to draw our own conclusions from a primary source (e.g. program "A" is better than program "B" because...) there is nothing wrong with using a primary source to verify that program "X" supports feature "Y". See WP:SELFPUB and Wikipedia:Identifying reliable sources#Self-published and questionable sources as sources on themselves for more on this. A program's own documentation, website, source code, etc will also almost always be more authoritative in documenting a program's current features and functionality than a 3rd party book which covers a software program because software evolves much at a much faster rate than books can be published.

Keep in mind, however, that the verifiability policy does not state that every single fact needs an inline citation. It would be quite silly and even disruptive to the reading of an article to require an inline citation for every single fact contained within an article. Anything controversial, suspect, or otherwise likely to be challenged by the average reader should however have a citation.

"... and inappropriately redirected to this article"? The Alternatives to deletion section of the deletion policy states among other relevant things: "Articles that are short and unlikely to be expanded could be merged into larger articles or lists."

71.214.52.97, I think *sshattery is a good way to describe it. The good news I suppose is the original individual responsible for harassing me online for roughly 18 months (who took an interest in harming articles such as this and others where I had previously done significant amounts of editing) is now indef blocked here on Wikipedia (although most likely back again editing under a new account). The two accounts which were "helping" him (including repeatedly mass-nominating this and many other articles which I had contributed to for deletion, mass-MFDing draft articles in my userspace, etc) were finally sanctioned after yet another lengthy ANI discussion and ArbCom amendment, after which both accounts were apparently "abandoned".

There are a number of highly skilled editors who work on comparison articles within WP:COMP's scope, so if I don't get to this one soon enough, eventually another editor will likely begin working on it. Comparison articles can be quite difficult to work on because of the complex markup and the sheer volume of information which has to be verified. I would estimate that cleaning the mess up here will likely take at least 20-30 hours of work, based on past experience. --Tothwolf (talk) 10:51, 10 February 2011 (UTC)

For this discussion to stop coming up repeatedly, just state the inclusion criteria somewhere like the top of the article and/or in hidden comments as recommended in, say WP:LIST#Lead section or paragraph or similar. This article certainly looked to me like most of the entries have a) an article and/or b) reliable secondary sources, for example, Chatzilla. Orion has neither. Your fight's not with me, and I'm not a deletionist, nor a campaigner against lists, comparisons, or you. If you reread what I wrote carefully, Orion doesn't have an article because it doesn't have RS. I can see GNG per se being tough for software to meet, but getting one or two secondary RS just isn't that high of a burden to meet. All I wanted was one or two RS mentioning Orion, to go forward with its inclusion here. Without RS, it really shouldn't, since nearly every other item in this comparison has RS. (Also note that I didn't delete it, just commented it out, because I respect the work involved in adding entries). Was I wrong to infer inclusion criteria based on extant entries? --Lexein (talk) 14:39, 10 February 2011 (UTC)

Example proposed inclusion criteria for this comparison:

1. Inclusion: One notable independent secondary reliable source: a review, or a two-sentence mention. Examples: Wired or Wired blog, Linux Journal, New York Times or NYT blog, or "industry" blogs which have been republished in or cited in notable publications. If the software author(s) is/are notable (established previously in sources like (1), example: JWZ), then the software has inherited notability.
2. Features/details: The primary source (software author's website) is good enough for the detailed software features. Detailed features do not require secondary RS for purpose of listing here. (In my opinion, writing about the detailed features does require secondary RS, because Wikipedia's "voice" is the voice of the sources not the editors.)

At the moment, criteria #1 is a barrier to Orion, but that barrier lifts as soon as it's noticed and written about. Wikipedia has always been about what secondary sources, say about the article subject.

--Lexein (talk) 01:42, 11 February 2011 (UTC)

Per WP:SELFPUB a self-published primary source can be a reliable source for non-controversial facts about the subject itself. As I mentioned above, a primary source can be used to show that "program X" supports "feature Y", and this is an acceptable use of a primary source here on Wikipedia. It is very common to cite a program's own site and documentation for verification of the release date/version and for features/functionality of the software itself. As I mentioned above, we can't use primary sources for other purposes such as "Program A is better than Program B because it supports feature X" because that would fall afoul of WP:NOR.

The notability argument for inclusion of content however has no place within a comparison article. Per WP:NNC we do not limit article content based on the notability guideline. The notability guideline covers whether or not a topic meets the inclusion criteria for a stand-alone article. What came out of our last RFC on notability and lists also extends this to the subject of a list, e.g. the subject (but not the title) of "University of X alumni who Y" would itself need to be considered notable as a group in order to have a "List of University of X alumni who Y" standalone list. As I've previously stated above, this seems to make perfect sense when it comes to such lists. What came out of the RFC still does not cover embedded lists, nor does it cover other forms of lists or articles.

Now, with this specific article, my argument is not that we should include every single IRC client ever written... While there are a finite number of "IRC clients" out there, I don't feel that we should include every pet project someone has ever written in Visual Basic (I often found that many of these were written in Visual Basic, however there have been popular software programs written in Visual Basic in the past too). We should however mention the features and functionality of popular or better known clients (as known by users of IRC) even when a client does not have a standalone article, regardless of whether or not it meets the requirements of the notability guideline for the purposes of a standalone article. E.g. XiRCON (see the AfD for that one and take particular note of who the nom was).

It isn't just here that this "discussion" keeps coming up. I've see these same notability arguments brought up on many comparison article talk pages and even when challenged and eventually struck down by others, they end up popping up again elsewhere. If we were to attempt to artificially limit comparison articles such as this solely to Wikipedia-notable items, we end up presenting a very skewed view of the overall subject. Comparison articles, especially software comparisons are among some of our most popular articles here on Wikipedia. See WP:COMP/PP for many such examples. While Wikipedia shouldn't be used for promotion of "software X", if a particular software program is or has previously been well known and/or popular with users of that genre of software, then we should still include it in such comparison articles.

The main issue I see looking at the larger view is that we currently do not have a Manual of Style page which covers the various forms of comparison articles. This leads to many people not having a fundamental understanding of how and why they are structured the way they are, why they include what they include, etc. Editors who work on these forms of articles tend to learn this stuff hands-on, but at some point we still really need to write a MoS page which covers these forms of articles. --Tothwolf (talk) 10:38, 11 February 2011 (UTC)

I had to WP:DISENGAGE for a while before responding to this.

1. Are you saying that Orion should be included in this comparison, even though nobody anywhere has written about it? No independent sources needed? That any joker in a garage can write whatever they want on their publicity brochure, website, blog, forum, and wiki, and you'd include it in this article, comparing its unverified (and UNVERIFIABLE) claims against other clients for which RS have established verifiability? That's a remarkably low threshold of inclusion, with which I disagree, and I'm an inclusionist. Keep in mind, there's not one shred of independent reliable evidence which supports the existence of Orion as an IRC client. By what Wikipedia policy/guideline/essay do you really think Orion's inclusion should stand? Ignore all rules - is this IRC client worth invoking that?
2. Linking Orion (blue link) elevates its apparent stature to "as mentioned on Wikipedia", purporting to link to an article with reliable, independent sources supporting Orion. Linking Orion to this very article is disingenuous, because this article cites no such sources about Orion. This constitutes a use of WP as PROMO, and should not be encouraged at all.
3. Notability is important as a concept, even though we're not talking about "article level" notability. We can't just make stuff up, or quote primary sources with zero independent verification. Comparison article or not, verifiability, WP:TRUTH, and WP:RS.
4. In case anyone wonders, I have nothing against Orion or its author. It has become the lightning rod, unfortunately, in this discussion, because it completely lacks any support in independent reliable sources, and therefore fails WP:RS, a bedrock policy at WP.
5. And again, to put a button on it, if you don't want discussions such as this to keep recurring, state the actual inclusion criteria concisely, at least at the top of the Talk page. Don't bother yelling at or insulting me, due to your failure to clearly state whatever consensus your prior discussions have reached. --Lexein (talk) 16:38, 5 March 2011 (UTC)

Addendum - Tothwolf, you state you only want to include popular IRC clients. Is there any RS way to measure popularity in terms of downloads or usage? Yes, Google shows there are over 17,000 "hits", but it's only mentioned as an echo of Tellini's own description of the software (using the word "comfortable"). I don't recommend relying on Softonic's download popularity rating (3rd of 19 IRC clients at 1000+ downloads), with Yahoo India's IM client is #1 at 143,000. BTW ChatZilla is at over 5,000,000+ downloads. Are there any usage-based server statistics reported in WP:RS which will indicate how many people are using which client? If not, how can we, as Wikipedia editors, decide download or usage popularity at all? We can't. We have to rely on WP:RS, because we can't just make stuff up. I say, in the absence of popularity reports in WP:RS, we have to abandon the notion of popularity in toto, and rely only on WP:RS verification of the existence, and hence, relative notability (not GNG article-grade notability), of the client.

Reminders: I didn't delete Orion, just commented it out until it has one WP:RS, ffs. I do respect the work. (Nobody's attacking anybody here except for you and an anonymous IP calling me an asshat. I politely request that you strikethrough or delete your rant against "attackers" - it has no place in a discussion of this article's improvement. As soon as you do that, I'll do the same for this. As for XIRCON, the deletion discussion was damnably short, and its undeletion request was denied but did you, really, make an effort to find RS to justify the article? There are some decently citable sources about XIRCON - I found them in 3 seconds. If you felt that strongly about XIRCON, why did you request to delete its userfied version at User:Tothwolf/XIRCON? If you don't care enough to maintain it for improvement in your userspace (dooming its invested work forever), you can't claim to care enough to bring it up here as an example of an "attack" on you or this article. Anyways, I'm now mindful of your experience at the hands of other editors, so I'll try to be less abrasive. But dude, knock off the victim/attack paranoia - it's just not happening here.)

Anybody else? Opinion about requiring one cloth-eared reliable source for inclusion? --Lexein (talk) 08:01, 9 March 2011 (UTC)

Lexein, you appear to be misinterpreting what 71.214.52.97 and myself were referring to as "asshattery", which were some of the "discussions" much further up on this talk page.

What I said above is Wikipedia-notability is not a valid metric for article content inclusion/exclusion. This is made clear in the notability guideline itself and I further explained why attempting to use Wikipedia's notability guideline to artificially limit article content ends up presenting a topic such as this to the reader in a biased "Wikipedia-notable" manner.

To further expand on what I said above, both primary and self-published sources can be used on Wikipedia and can be reliable; it depends on the source and how it is used. While blogs, email lists, etc are generally not considered "reliable" (especially for the purposes of a BLP article), per WP:SELFPUB and longstanding common practice within WP:COMP's scope, we can and do use primary and self-published sources in the manner I described above. Take for example the Linux kernel. Emails, usenet posts, and blog posts made by Linus Torvalds and other Linux kernel developers are considered reliable for the purposes of citing "The Linux kernel supports feature "X"" and "The next version of the Linux kernel will include support for "Y"". While we certainly can't use them to say "The Linux kernel is better than [insert operating system here] because ..." (since that would be considered original research), if a developer made such a statement on his blog, we could still include his statement as a direct quote.

As for Orion, the software program's own site [1] for example is reliable for the purposes of showing that the client is released under the Creative Commons Attribution-NonCommercial-ShareAlike license. The source code [2] and included documentation (Changes.txt, Readme.txt, etc) would also be reliable for the purposes of showing that the software supports features "X", "Y", and "Z". While English sources are much preferable (see WP:RSUE), if we don't have an English source which states that open source software program "X" supports feature "Y", the source code of the program itself can be cited as it is a reliable self-published source and is acceptable for the purposes of verifiability when it comes to the software program's features and functionality.

As far as defining popular IRC clients, yes, there are metrics we can use, although two of the better online ones such as the "Top IRC clients" from irc.netsplit.de and another from Alexa are no longer available. This really isn't surprising either, given the slow decline in popularity of IRC with the introduction of some of the newer web-based social networking technologies.

Yes, the XiRCON AfD was very much flawed, but it is just one example of many which were part of a much larger pattern. The reason I CSD'd all drafts within my userspace is detailed in the AN/I report I made here. After I made this report, one if the individuals involved attempted to initiate an ArbCom amendment request [3] in revenge (which Plaxico'd quite badly). Two of the three individuals involved abandoned their accounts when they were finally sanctioned and placed under edit restrictions. The third individual was later indef blocked for continuing to make personal attacks. For more background, see User talk:Tothwolf/Archive 4#On the subject of COI. These three individuals were the ones involved in the discussions above which 71.214.52.97 and I referred to.

Another example is Wikipedia:Articles for deletion/AmIRC. Here are two dead-tree sources for it:
Charalabidis, Alex (1999-12-15). "Odds and Ends: IRC for Other Platforms: Amiga". The Book of IRC: The Ultimate Guide to Internet Relay Chat (1st ed.). San Francisco, California: No Starch Press. p. 249. ISBN 1-886411-29-8. AmIRC is probably the most popular client for the Amiga. It supports all the basics, emphasizing well-developed standard features rather than modern toys, and provides a stable and secure chat environment.
Jeacle, Karl (1996). "Running the Best Software: Internet Relay Chat: AmIRC". First Steps Amiga Surfin'. Bedford, Bedfordshire: Bookmark Publishing. p. 76. ISBN 1-85550-007-8. AmIRC is the aforementioned competition to Grapevine. As with the other "Am" applications, it is the Amiga Technologies choice for inclusion in the A1200 Surfer pack. {{cite book}}: Unknown parameter |month= ignored (help)

Another example is Grapevine, and while I can't see that we've ever had an article for it on Wikipedia (and it would seem to meet the requirements of the GNG for the purposes of a standalone article), coverage of this client certainly should be included in this article.
Charalabidis, Alex (1999-12-15). "Odds and Ends: IRC for Other Platforms: Amiga". The Book of IRC: The Ultimate Guide to Internet Relay Chat (1st ed.). San Francisco, California: No Starch Press. pp. 249–250. ISBN 1-886411-29-8.
Jeacle, Karl (1996). "Running the Best Software: Internet Relay Chat: Grapevine". First Steps Amiga Surfin'. Bedford, Bedfordshire: Bookmark Publishing. p. 75. ISBN 1-85550-007-8. Grapevine has been called "The best IRC client on any platform" by a number of experienced Amiga Internet users. it was one of the first major Internet applications available for the Amiga and it did such a good job of being an IRC client that until recently there was no competition whatsoever. {{cite book}}: Unknown parameter |month= ignored (help)

I could cite dozens and dozens of these but I think you probably get the idea. Note that The Book of IRC like many books is not indexed (either full or in part) and text-searchable with Google, so a simple Google Books search is not going to turn it up as a source for the great many IRC-related subtopics it covers. (cf. FUTON bias, Wikipedia:Reliable sources/Cost, and Wikipedia:Perennial proposals#Require or prefer free, online sources) --Tothwolf (talk) 10:23, 9 March 2011 (UTC)

About IRC clients which have independent RS, I of course welcome including them here. Old sources are fine. Dead tree sources are fine. Clients with zero independent WP:RS should wait. Including such clients in a survey or comparison article is solid Original research and/or WP:PROMO. Let me be clear: there's article-level notability (GNG) and sentence-level notability (RS); I've only been advocating sentence-level notability (I've never claimed that GNG applies within this article, and I can't understand why you claim I did. Carefully reread what I wrote). Solution: if some Random Tech Journal mentions Orion in a survey of clients, mentions one or two features, and comments on whether it's "comfortable" as claimed by the author, that would clear my ever-so-low entry hurdle. This comparison article can then certainly list the rest of the features. I'm fine with that. Finally, perhaps I wasn't clear, but not even bloggers or forum folk have reviewed Orion. Nobody. That's how unabashedly non-notable it is. I doubt that any of the other IRC clents listed here can make that claim. --Lexein (talk) 13:36, 10 March 2011 (UTC)

Based on your reply it appears you at least understand most of the points I tried to make.

One important point which still doesn't seem to be translating is the threshold for inclusion of content in an article is verifiability, not notability (or sentence-level notability as you refer to it). A reliable source, be it self-published or independent third-party which allows for verification is what is required. Your earlier argument was that we couldn't even verify that Orion existed -- the link to the website and source code for the program certainly shows that it exists. It also shows that the program is a freely available open source software program made available under the Creative Commons license. One thing (among others) which makes Orion quite noteworthy within the larger IRC community is that it is only client of its kind which is made freely available under the Creative Commons license. This alone makes it worth including here for readers who are searching for a freely available open source software IRC client for the Microsoft Windows platform.

Given your sentence-level notability argument, would you argue that because mIRC's latest features and functionality have not been covered in a dead-tree (cloth-eared?) or otherwise independent third-party source that we cannot mention them here? How about in the mIRC article itself? We can certainly verify those features using mIRC's own documentation and website, which are certainly reliable and acceptable per WP:SELFPUB, but you won't find much in the way of "reliable" independent sources which give it too much coverage. You may find a number of people discussing the latest features in online forums or on their blogs, but those are far less "reliable" than the program's own site and documentation. As I said earlier: "A program's own documentation, website, source code, etc will also almost always be more authoritative in documenting a program's current features and functionality than a 3rd party book which covers a software program because software evolves much at a much faster rate than books can be published."

Your other argument of WP:SPAM (WP:PROMO) does not apply to the inclusion of Orion here either. Orion is a freely available open-source software program (verifiable). The developer does not make money from making the program freely available, and it actually costs him time (development, documentation, support, etc) and money (hosting) just to give it away to the larger online community. On the flip side of the coin, if this was instead a little-known commercial software program which a developer or company sold and used as a revenue source, I would have removed from this article myself, as I have with other software in the past.

The developer for Orion also didn't coverage of it to this article, I did, back in June 2009. While I don't think you intended to imply that I'm somehow a spammer for adding it here, that is exactly how it comes across to others when you referenced the WP:PROMO section of WP:SPAM.

Under the notability guideline, Orion doesn't (at least yet) qualify for a stand-alone article. As I stated above, the Alternatives to deletion section of the deletion policy states among other relevant things: "Articles that are short and unlikely to be expanded could be merged into larger articles or lists." This is also echoed by the WP:PRESERVE section of the editing policy and the WP:BEFORE section of articles for deletion. Per these policies and practices, Orion (IRC client) (which at the time was a tiny little sub-stub [4]) was merged and redirected here, where we were also able to give it much better coverage. --Tothwolf (talk) 18:11, 10 March 2011 (UTC)

Your WP:UNCIVIL snark about my level of understanding notwithstanding, Wikipedia isn't about primary sources. Orion has zero verifiability: the fact of its existence is based entirely, and endlessly, on copy/paste of text from its original website (which lists zero details), and a few non-RS distribution sites. That's one way promo occurs - no matter who's doing it. The continued inclusion of WP:PROMO reflects badly on every Wikipedia editor who touches an article, and on Wikipedia as a whole. At no point have I implied any bad faith on any editor's part here - promo can be unintentional.

Everything about Orion in these tables is unsourced, and at the moment, unsourceable - therefore, it's WP:OR. Nothing added was based on reliable third-party sources, or the author's Orion webpage or Orion goes Open Source blog entry, or the program itself. The only claim supported is that it runs on "Windows", and that's on a separate page entirely. Where's the source that it doesn't handle proxies? The source code? The GUI? This is not a wikiwiki about software; verifiability at Wikipedia is intended for general readers to be able to verify claims made in any kind of Wikipedia article (prose, table, list) in reliable sources. Explicitly, WP:PRIMARY states, as policy, "A primary source may only be used on Wikipedia to make straightforward, descriptive statements that any educated person, with access to the source but without specialist knowledge, will be able to verify are supported by the source." The "specialist knowledge" clause would tend to prevent using program source code as a primary source for lists of features, but may allow facts from "right-click/properties", or "Help" files.

There's a basic problem with using primary sources in comparisons: tables. In prose, Wikipedia distances itself, as described in WP:THIRDPARTY#Non-independent_sources as the authority for claims, by using language like, "The author of the software claims (x)." But in this article, in this table, there's no opportunity for this qualifying language. The presence of the table format itself lends WP:UNDUE weight to the table contents, as if they are verified as accurate, when in fact, no such proof exists, not even in the author's website. Do you really want Wikipedia to lend an authoritative "voice" to fundamentally unverified factoids? Yes, editors can "verify" features themselves by running the software, but that's original research, and it's not the way it's done at encyclopedias, or here.

Alternatives to deletion is a dead horse in this discussion - I daresay it is intended to apply to articles with prose, not lists or tables of comparison, and is not intended to give a pass to material lacking any independent reliable sources, only, as it explicitly states: "articles which are short and unlikely to be expanded." It is not appropriate to use WP:ATD to shoehorn a non-notable, non-verifiable item (Orion) into a larger encompassing article which contains no prose, and into a comparison table consisting of mostly independently WP:RS, verifiable items, mostly possessing their own articles.

mIRC has coverage in RS, both in magazines and books. What are you on about? As soon as a topic has any coverage in RS, a primary source can supply additional details about itself. Which has been my point all along. But here, Orion's publicly verifiable (verifiable by non-programmers) facts amount to nothing.

WP:OTHERCRAPEXISTS. Other articles about software do in part rely on primary sources such as websites, changelogs, even software binaries themselves, for occasional facts. These inclusions are always on the bubble, because they live in the shadow lands of not supported by reliable third party sources, and frequently not supported by primary sources. At any moment, a deletionist may challenge unsourced material, and delete it. For the sake of this article's strength in the face of AfD, it should not include unsourced material.

What's strange here is the inclusion of essentially abandonware which hasn't changed or been developed by anyone since 2006, which was never notable, and is now unlikely to ever be notable, or even (be honest) verifiable except through original research. On the website, Simone Tellini writes, "Since I haven’t had time to dedicate to this project in years, it’s now available under a Creative Commons license...". Specifically, per the blog, since 2006.

If Orion is so great, why hasn't even one independent reliable source written one thing about it?

Since this is only you and I on this, and since there are no clearly stated inclusion criteria, and since you seem unwilling to see shoehorning of unverifiable, unsourced claims as a problem, I think the next logical step is WP:3O.

--Lexein (talk) 08:47, 15 March 2011 (UTC)

I agree with Lexein on this. (see below) IRWolfie- (talk) 20:43, 21 March 2011 (UTC)

Lexein, I wasn't trying to be snarky with you, however if you are really only here to try to prove a point and just want to wikilawyer, I really don't feel that there is much of a reason for me to continue to attempt to have a discussion with you. Your comments thus far seem to indicate that you have a general dislike of Comparison of articles. Honestly, I'm not sure why you have chosen to focus on this specific article if it offends you so. There are millions of other articles you can go work on if you really don't like the content of this one...

Lexein, I really don't mind civil discourse and discussion as to the hows and whys "Comparison of" articles generally include the types of material they include and are structured the way they are. I understand not everyone (myself included) is going to be familiar with every single style and convention used on Wikipedia. That said, while you are certainly entitled to your opinions, and I've been more than willing to hear you out, it still doesn't change the fact that some of your arguments such as "Orion has zero verifiability" are factually incorrect.

Orion verifiably exists. I can't say the same thing for a number of other clients I've previously removed from the tables here which appeared to be drive-by additions. Many of these I couldn't find anything about them, be it primary sources, secondary sources, or really anything else. I'm not sure what you are referring to as a "copy/paste of text" as nothing was copy/pasted from somewhere else.

Orion's basic features and use of the Creative Commons license as an open source software program for the Microsoft Windows platform is verifiable using reliable (primary) sources which includes both its own website and the license text embedded and included with the source code for the program itself. Your apparent personal dislike of certain types of primary sources notwithstanding, material simply can't be WP:OR if it can be verified.

Further, not everything requires a citation (in particular, "Subject-specific common knowledge"), although as I've previously stated, I much prefer to have something to show that a particular program verifiably exists. I also really like to see an inline citation for the latest release version and date for a software program because it makes it much easier to keep an article updated. A non-contentious fact such as a native Microsoft Windows program not being supported by Mac OS generally does not need a citation. Having table cells which indicate this simply allow for sortable tables and makes the material more accessible to readers. The reverse however, might be best served by one or more inline citations, as it can be somewhat unusual for a native Microsoft Windows program to also have a native Mac OS version. That said, it wouldn't be all that surprising to see a well-written Unix-like program that is also compatible with say Mac OS X that can also be compiled under Cygwin on the Microsoft Windows platform.

The alternatives to deletion section of the deletion policy applies to all content, not just prose. It does not matter if you personally consider something to be "non-notable" or not. If our readers find the content you consider "non-notable" useful, we still have an obligation to include it. Despite your comments otherwise, general readers have demonstrated time and time again that they wish to be able to readily find more information in these comparison articles about software programs such as Orion by making comments to this effect on these talk pages and by trying to add such information to the article tables themselves. There also appears to be a very strong interest in open source software for the Microsoft Windows platform. Consensus outside of the discussion you and I have going on here continues to be that "Comparison of" articles are both encyclopedic and helpful for our readers. It has also been previously demonstrated that the more information these articles contain (subject to usability limits for typical web browsers), the more helpful our readers find them.

WP:PRIMARY does not state that the language of the source has to be English only, and as stated above, this is covered in WP:RSUE section of the verifiability policy. An "educated person" who can read and comprehend the language the source code is written in does not need to have any sort of "specialist knowledge". The ability to read a programming language is no different from the ability to read another foreign language. In fact, in most cases, the program's own documentation, which you also refer to as "right-click/properties" and "Help files" is more than sufficient anyway.

As for "qualifying language" we can and do use footnotes, which when used with tables, are typically located directly below the table. [5] There was previously work underway right here within this article by a number of editors (including myself) to add additional footnotes and citations. The first step which was previously completed (before another editor came along and blanked roughly half of this article, which included the removal of sources in some cases) was to verify the existing content of the article and to correct or remove anything inaccurate or which could not be verified. This improvement work stopped when the events I linked to above began. At that point in the revision history of this article however, the material presented in the article was verifiable and had been fact checked by a number of editors.

The latest/current features and functionality of mIRC have not been covered in anything other than primary sources, including the program's own documentation. This is true for most software programs. You can't argue it both ways. Self-published primary sources can be used for such features/functionality verification and the notability guideline does not apply to article content.

I really don't see the point in involving WP:3O just because you don't personally agree with the longstanding practice of using reliable primary sources to verify software features and functionality for computing articles. This has been discussed previously within WP:COMP and WP:RSN and while you might not like the fact that we can and do cite a program's own documentation and/or source code for non-contentious facts about the program itself per the WP:SELFPUB section of the verifiability policy, this is commonly done and there is little point in arguing about it here.

To sum this up, again, the notability guideline does not restrict or limit article content. Self-published sources are perfectly acceptable for the verification of features and functionality for a particular software program. While English sources are generally preferable, it doesn't matter which language a source is written in – if a non-English source is available when an English source is not, or the non-English source is better, the non-English source can and should be used.

Lexein, there actually are other editors lurking and reading our discussion [6] but I daresay that the majority just don't want or have time to argue these complex points with you :) --Tothwolf (talk) 06:35, 22 March 2011 (UTC)

Many of the clients listed here are present with WP:UNDUE weight. "Views that are held by a tiny minority should not be represented except in articles devoted to those views." IRWolfie- (talk)

IRWolfie-, I don't agree with your WP:UNDUE assessment. We aren't attempting to discuss a "minority viewpoint" and the content is nothing more than facts, which is about as neutral as it gets. This article is specifically limited in scope to IRC clients, of which there are a finite number. There are roughly less than 100 client programs that have actually been popular with users over the last few decades (Microsoft Windows isn't the only operating system...), which prior to the blanking that occurred here, we only covered roughly 55-60 or so of those. Because we are already working within a very limited subtopic, claiming coverage of clients violates WP:UNDUE makes little sense.

As I've said previously, we end up violating WP:NPOV by presenting a skewed view of the topic as a whole if we attempt to misuse the notability guideline in order to remove and artificially limit article content to solely wiki-notable material. This is part of why the WP:NNC section of the notability guideline exists. If you misuse the notability guideline to remove and limit article content in an article such as this, it effectively turns a comparison article into nothing more than a navigational list or directory of other Wikipedia content, in which case, what is the point in even having comparison articles?

Subtopic navigation is not the intended purpose of a comparison article. Navigational lists are quite different from comparison articles and both are helpful for different purposes. While I do agree with you that we need some way to exclude some of the really "unpopular" (which is still a subjective way to describe it) clients (like the random visual basic clients I mentioned that anyone can whip up fairly quickly), we simply can't do this with the notability guideline as it was never intended to do this and is ill-suited for the task.

A fair number of entries which were removed from this article during the mass blanking are covered in books which are not indexed by Google Books (which I mentioned in earlier discussions). As I've also already mentioned above, Google Books does not index all works which give coverage to this subtopic or to the parent "Internet Relay Chat" topic. Google and Google Books are not magic and should not be used as a sole means of information gathering. The parent topic dates to about 1988, which predates the World Wide Web (1991), Google (1996-1998), and the Wayback Machine (1996 or thereabouts). This combined with the highly dynamic nature of web content (here one day, gone the next) means it can be just about impossible to use Google to find secondary sources for some of these older clients, even though they were once extremely popular and were at one time discussed widely. Can you find a non-primary source for the very first IRC client (simply called "irc") which was initially distributed with IRCd? I've been lucky just to be able to find the source code for this historic software program. Using subjective importance in terms of "I've never heard of it, it isn't notable" is not valid justification for removing article content.

For someone who is unfamiliar with the larger history of IRC, many of the clients might seem obscure, however we should give them limited coverage where we can (such as in a comparison article) so that people who wish to learn more about the larger topic can find more information on them. WP:ITSLOCAL states: "Stating an article should be deleted because you and most of the world do not know about it is akin to the I've never heard of it argument. Many subjects are esoteric, meaning that only a small crowd is familiar with them. For example, few people are aware or interested in some obscure forms of living things, space bodies, or scientific concepts, and few people will ever know about them in the first place in order to even desire to read about them. Yet there is sourced information about them, so they qualify to be included." and "The same is true about subjects only of interest to those in a single city, town, or region. People who live outside the area who have never visited there or done any research on the area will obviously be unlikely to have ever heard of them. But Wikipedia is not limited to subjects that everyone in the world knows or will have a good chance of knowing. Being a global encyclopedia, Wikipedia can cover a wide range of topics, many of them pertaining to the culture of a single country, language, or an ethnic group living in one part of the world. The people living in a single city or town and everything they have built around them are likewise a culture and society of their own." This becomes even more pertinent when you consider that Internet Relay Chat is effectively a society of its own (a fact which you can find stated in published works).

While the notability guideline should not be misused to limit the content of an article, notability is also not temporary. Just because someone was able to mass-prod and mass-AfD many of these articles (with about 1/3 or so still not yet restored) without due diligence and as an intentional means of being disruptive does not mean they should not have standalone articles. I've had many restored and worked on them one at a time as time permits, but the cleanup process for that mess is much more work than it was for that individual to mass-prod and mass-AfD articles.

IRWolfie-, since you seem to have an interest in Internet Relay Chat, if I compile my notes and create some sandboxed outlines and drafts, would you be interested in helping to sort this mess out and rebuild this content to make it more useful for our readers? How about articles for clients such as WSIRC, [7] which was the dominant Microsoft Windows client prior to the introduction of mIRC? How about Grapevine for the Amiga platform, as noted above? --Tothwolf (talk) 00:10, 23 March 2011 (UTC)

I was very explicit in saying that I think they have WP:UNDUE weight for inclusion, I did not mention notability. Undue Makes sense because some like ii which don't have their own article should not be included as the above quote showed. I should note also you reverted my edits where I removed ii because it was 'notable'[8]; as you have just mentioned this is not grounds for inclusion and all others should also be removed that. (Side note: I notice you and Lexein leave giant comments, is this not a bit much!?) IRWolfie- (talk) 01:12, 23 March 2011 (UTC)

IRWolfie-, I didn't revert you specifically. If you check the edit history of the article, ii was included here previously and was one of those which was part of the mass-blanking of this article. I combined what an editor tried to add with the version I had in an offline copy and merged that back into the current version of the article after fact checking it (I found and corrected one minor mistake in the process).

ii is as far as I've been able to determine, the only IRC client which uses a FIFO filesystem as the "user interface". This makes it unique enough where I feel it should be included here. I'm not able to locate enough with a quick search where I feel it would meet the requirements of the notability guideline for the purposes of a standalone article, but it has been covered in depth by IRC-Junkie.org, [9] which is considered a reliable secondary source for IRC-related news and information. It has also been a topic of interest over at reddit. [10]

WP:UNDUE was never intended to exclude something like ii from an article such as this. There is no requirement that a subject have a standalone article (per the notability guideline) before it can also be discussed within another article. If this was the case, Wikipedia would be much, much smaller... As previously stated, the Alternatives to deletion section of the deletion policy states among other relevant things: "Articles that are short and unlikely to be expanded could be merged into larger articles or lists." The WP:PRESERVE section of the editing policy and the WP:BEFORE section of articles for deletion also cover this.

Would you argue WP:UNDUE in the case of the very first IRC client which was called irc that I mentioned above, or WSIRC, or even Grapevine? I agree that there are some things we should exclude here, however I'm of the opinion that we are far from violating WP:UNDUE by including information which is very much on-topic and of interest to readers who would otherwise be interested in reading about this (generally boring) subject. --Tothwolf (talk) 03:39, 23 March 2011 (UTC)

Your grounds for including ii is that you think it is unique, uniqueness isn't grounds for inclusion either. you're basically saying it's notable just without using the word 'notable'. Being mentioned in irc-junkie isn't grounds for inclusion :). IRWolfie- (talk) 09:45, 24 March 2011 (UTC)

LimeChat

There are two different IRC clients named LimeChat. One is the one mentioned in this article, the other one is a Windows client by the same author. See here: http://limechat.net/ . It's currently at version 2.39, and seems to be available in Japanese only. It's, however, the most used IRC client in Japan, or at least as far as I remember, all Japanese people I've seen or known, use it. There's also a third one, for iOS. - 94.140.73.150 (talk) 11:43, 1 March 2011 (UTC)

Please provide reliable sources to show it is the most used in Japan. IRWolfie- (talk) 16:40, 3 March 2011 (UTC)

Firm rules for inclusion needed

Firm rules are needed for what should and should not be mentioned. The existence of a wiki article is a good indicator since it indicates reliable sources discuss the client. If the article doesn't exist, create one! otherwise If it's not notable enough for an article then it probably shouldn't be included here. ii is a good example, I've never seen a reliable source discussing it. IRWolfie- (talk) 20:35, 21 March 2011 (UTC)

I agree. It's time to establish a firm criteria for inclusion. The requirement of having an article looks good enough.

If one of the clients is not notable, then its article can be challenged via PROD and AFD (and contested at DRV). --Enric Naval (talk) 13:51, 24 March 2011 (UTC)

The problem is that notability != importance. I'd rather go with the idea that if there is coverage of a client in reliable secondary sources, whether or not it would meet the GNG, and if the details which are needed for the the comparison can be reliably sourced, then it is worth including. The more inclusive the list, the more valuable it is to readers - and while there needs to be a limit that stops just anything being added, the GNG is too far in the other direction. - Bilby (talk) 13:57, 24 March 2011 (UTC)

Right, my point has been that notability in terms of Wikipedia's notability guideline just doesn't work well for article content (hence why it explicitly states that the notability guideline does not apply to the inclusion/exclusion of content itself).

The problem with mandating a secondary source before adding material to this article where we already have reliable primary sources for plain and simple facts is that we still end up limiting the article's content in a subjective way and excluding material which is of interest to readers. As I mentioned above, due to the dynamic nature of the internet and web, you won't be able to find a non-primary source for anything related to many early (and previously popular) clients including the first one which was called "irc" that shipped with IRCd.

One major problem I've noticed in these discussions is that there are a number of editors who for whatever reason either dislike or don't understand that primary sources can be reliable in cases such as this where all we need them for are simple facts about the software itself. We've used such sources for software as far back as I know and they have never been a problem when used in this way.

I'm really beginning to question why we are even having this discussion here since we have not really had any sort of problems with spammers and drive-by promotional additions. This article used to be very "quiet" and fairly easy to maintain. There were a handful of editors (including myself) who did the majority of the cleanup and maintenance on this article. Had I not been involved with editing here, the editor who was keen on disrupting anything I edited as part of a larger harassment pattern (who is now indef blocked) along with a couple of his "friends" would have never even started this debate here in the first place. (There are also a few more questions directly related to the past harassment pattern which are beginning to form, but unless it becomes a problem again I don't see the need to bring those up at a larger venue.) --Tothwolf (talk) 17:40, 24 March 2011 (UTC)

"Popular" again, eh? Produce a reliable source supporting the popularity of any these obscure, decidedly unpopular clients. Wikipedia is not about sources you like, it's about independent reliable sources. Perhaps you missed reading WP:ATA - arguments to avoid in deletion discussions such as "It's interesting" and "it's popular."

I'm calling bullshit on your misrepresentation of my words. I never said primary sources can't be reliable. Primary sources can only be reliable about themselves as far as they actually make statements about themselves. Orion's author has written little or nothing about Orion - I've asked you to produce evidence of any such writings. None of the claims about Orion in this table, are sourced from anywhere, primary or third party, and definitely not the author's website or documentation.

Primary sources can be used in Wikipedia, if written as "Person says thing about themselves" or (stretching the meaning of primary source to the breaking point) "Software documentation lists this feature", with the source in an inline citation. Because this is not a wiki about software, and it is for general readers, it's NOT done to state "The software's source code claims this feature." In this comparison chart, (a) there's no place for any such disclaiming language, and this adds undue weight to the claims made, and (b) no inline citation is offered (with a primary or third-party source) which supports any of the claims made (and in the case of Orion, no such will likely ever be found). Sources, sources, sources. Why hasn't even the author written about its feature set, as detailed here, anywhere? And since the author hasn't written about its feature set (except in the source code), isn't anything you write about it an example of the very definition of original research? As Charlie Brooker has said, "Don't say it isn't, it is."

--Lexein (talk) 09:08, 25 March 2011 (UTC)

As I have previously mentioned undue weight is grounds for removal, "Views that are held by a tiny minority should not be represented except in articles devoted to those views.". When applying it to this comparison table it would be "Clients that are used by a tiny minority should not be represented except in articles devoted to those clients." Without significant discussion by reliable sources I see no reasons for inclusion, e.g being mentioned once by irc-junkie for example doesn't sound like significant grounds. Perhaps the unpopular clients and clients with reliable third party sources should be separated from the others so that they can stand or fall on their own weight? IRWolfie- (talk) 18:36, 25 March 2011 (UTC)

This isn't a deletion discussion, and this isn't about fringe views. Primary sources are useful for factual, uncontentious data: whether or not some software is written in C or Java, of if it runs under Windows or Linux, is exactly the sort of thing we can rely on primary sources for. Whether or not the software is extremely popular, or how it has influenced the market, is the sort of topic we can't rely on primary sources for, but that's the sort of question we don't ask in a comparative list like this. In regard to minority views, that policy was designed to reduce the impact of flat earthers or conspiracy theorists - it doesn't relate to the popularity of software, which also can't be calculated by coverage in reliable sources. I don't have a problem with the idea of having inclusion standards based on considerations of due weight, but applying those sorts of policies to a list about software won't really help things. - Bilby (talk) 23:04, 25 March 2011 (UTC)

So what then is the criteria for inclusion? What is to stop -any- and every client being added? IRWolfie- (talk) 23:10, 25 March 2011 (UTC)

IRWolfie-, To answer your earlier question first, detailed software reviews from a site that has a well established peer review system or the author of the review is a well established expert (such as what you see with the review of ii from IRC-Junkie) are considered to be reliable secondary sources, hence suitable for the purposes of establishing notability for the purposes of creating a standalone article. Not all software reviews are going to meet those qualifications though, including the canned reviews software authors can submit themselves to certain sites. In the past we only had a handful of sites like IRC-Junkie which were dedicated to IRC news, but today that is pretty much the only one left. IRC-Junkie has also previously pass muster for being a reliable source in and of itself, including several AfD discussions, so that review for ii is actually meaningful.

On to your current question, there actually aren't that many IRC clients out there. As I mentioned above, there are really only about 100 or so which have ever had much exposure at all (which covers all of these different operating systems and system architectures). That in and of itself defines a hard limit for what this article could ever possibly cover.

In the past I tended to remove drive-by entries which I thought were promotional efforts for commercial software, but I can count on one hand just how many of those I've had to deal with over the last 2+ years. We just never really had a problem here with drive-by additions trying to promote commercial software. The last several additions which have been discussed here were attempts by readers to restore content which this article previously covered before it was largely blanked by the efforts of a single individual.

We are also limited in terms of article size which is why both IM clients and mobile IRC clients are not included here. Article size limitations combined with the radically different operating system requirements of handheld/mobile devices are the reasons why mobile IRC clients were split out into their own comparison article. (Note that many of those clients are extremely popular with users of handheld devices and are very much in need of either standalone articles or coverage in a larger prose article.)

There is also common sense, in that if we can't find anything about a particular client, be it secondary coverage, primary coverage for non-controversial hard facts such as software requirements, features, functionality, etc, it will likely be considered either vaporware or a hoax and is unlikely to remain in the tables here.

The previous approach worked well enough which is why this massive discussion seems a little silly for this particular (boring) article. I really consider it a shame and a disservice for our readers that all this debate and disruption of the article content boils down to the previous disruptive actions by an individual who took to following my contribution history to find articles I previously edited in order to disrupt them and create strife among whoever they could draw in. Even though they are now no longer able to fan the flames on this talk page, the problems they created months and months prior continue. --Tothwolf (talk) 05:26, 26 March 2011 (UTC)

It's not about promotion, it's about cruft and about non-notable items making it into lists. As I was saying, we need a inclusion criteria: that all entries have an article. --Enric Naval (talk) 06:16, 26 March 2011 (UTC):

Enric Naval: An article may be too much to ask. One independent, third-party, reliable source, plus one primary source (for details listed in the table), would be more than sufficient. I'm being inclusive here, and inclusionist as well. I'd accept a blog entry from an industry-notable author (an author who has been published in other notable publications.

To clarify: this discussion is only about this article's merits. What's happening is that a number of editors believe the material here should be sourced, or not be here. WP:RS are best. Primary sources can be used if they actually write something about themselves. (Orion's author wrote nothing about Orion.) "Hard facts" don't count unless they are, in the main, sourced. Unsourced "hard facts" are the very definition of original research.

• The only definition of "coverage" that matters to Wikipedia is discussion in third-party independent reliable sources. Primary sources aren't "coverage."
• This article is being scrutinized and discussed because its inclusion criteria are non-existent, and much of its content is unsourced.
• Several editors, (not "a single editor") aside from me, have removed unsourced material (though I'm the only one commenting out rather than deleting).
• There seems to be confusion of "uncontroversial" with "unchallengeable". All unsourced material can be challenged, and I have done so with Orion: it's unsourced because you didn't cite an independent WP:RS source, and even the primary source has no details; further no source exists which mentions its details; hence, it is unsourced.

The sad thing, from a Wikipedia policy, guideline, and essay point of view, is that this article has been allowed to fester, unsourced and unweeded, for so long. It might be a lovely tall garden of unsourced weeds, but they're weeds, nonetheless.

Look, this is not difficult. Find the independent RS sources to support a substantial number of claims about a client before putting material about that client in the article. I'm really not that fussy, believe me: I'm an inclusionist. --Lexein (talk) 10:04, 26 March 2011 (UTC)

I would settle for multiple RS about the client, translating to "at least 2, better if it has 3". To avoid listing clients that only have one review in one publication, because wikipedia is not a directory listing of software programs. Borderline cases with one or two reviews can be discussed case-by-case, looking at the claims made by their RS.

For example, under this criteria, "ii" and "Orion" wouldn't have entries. But AmIRC would have an entry, but only because one of the porters claims that it influenced XChat (and, that source is not all that good; personally, honestly, I feel that it would be a bit weird not to have a single Amiga client in the list). --Enric Naval (talk) 11:36, 26 March 2011 (UTC)

Enric, "cruft" and "non-notable" are your own personal opinions. You already made it clear that you aren't too familiar with this topic or material, so it would stand to reason then that continuing to try to apply your own subjective criteria is fatally flawed. You even removed the entry for KVIrc [11] when that client had more than enough coverage (including published works) to justify a standalone article. The mass-prod and mass-AfD behaviour directly attributable to a small tag team which resulted in the majority of these "red link" entries was disruptive and problematic and it looks like the majority of us at least agree on that fact even though there is a lot of disagreement here with regards to this specific article. As for AmIRC, didn't you read above that we have two published sources for it anyway? Google is not a magic oracle.

Enric, the WP:NOTDIR argument you've tried repeatedly to use here hasn't held up in talk page discussions or AfDs for computing and software comparisons so you really should let that one drop.

Lexein and Enric, neither of you have addressed my point above regarding historically important clients like irc that originally shipped with IRCd. The only real sources we have for it today are primary sources such as its source code and the limited documentation. This also doesn't address clients such as XiRCON and Zircon which were also highly regarded by users and influential and previously included here.

Lexein, your personal opinion about both ii and Orion noted, I firmly believe both should be included here because our readers clearly want to know more about them. Wikipedia exists for our readers. Wikipedia is not a game for editors to play around with and keep busy with. How does removing simple factual data about these clients in any way help our readers learn more about this subject? Quite simply, it doesn't. Both of these specific clients are unique (as were most of those which were removed as part of the mass-blanking of this article) and our readers have demonstrated that they want to read more about them.

The size of the article was manageable, it was maintained, it had been fact checked (and work had been underway to expand the footnotes and in-line citations), and it did not have a problem with spammers using it for promotion of their commercial wares. Since we already violated the Wikipedia-equivalent of Godwin's law by bringing up conspiracy theorists above, to quote the late great Chuck Shramek, What the hell is going on?!

While I fully agree that we really should have sources (be it secondary or reliable primary), trying to limit this article's content in a subjective way, be it up front WP:N or backdoor "must have a reliable secondary source" (exact same result as WP:N) before we can have an entry here only hurts our readers. This whole subjective busy-body task of removing on-topic, verifiable material only results in an article which is of far less value to our readers. --Tothwolf (talk) 21:57, 26 March 2011 (UTC)

You are right, I hadn't spotted that part of the discussion. So, that only means that AmIRC would be guaranteed an entry in the list. It doesn't undermine my main point: that we should remove entries with no RS. --Enric Naval (talk) 19:42, 27 March 2011 (UTC)

I'll be brief. There is zero value to Wikipedia readers if claims cannot be verified from independent reliable sources. This is a fact of Wikipedia enshrined as a bedrock policy with which you have so far refused to embrace. Claims unsupported by published independent sources directly damage Wikipedia's credibility, reducing it to a blog, a catalog or PR. Once the toehold of even a single discussion by third-party reliable source is established, I'm willing to be inclusive of primary sources which actually write about themselves (which Orion's author has not). BTW I'd even cite newsgroup archives. Nothing about Orion there, either, I'm afraid. --Lexein (talk) 17:28, 27 March 2011 (UTC)

A vote (open for a week) should perhaps be taken to determine if an existing article on the client is needed for inclusion here or not. IRWolfie- (talk) 22:42, 26 March 2011 (UTC)

While I don't like making short replies with just a link to a policy or guideline, such a vote would be counter to WP:NOTDEMOCRACY and WP:VOTE. Aside from that, a one-size-fits-all approach isn't going to work, be it here, or with other "Comparison of ..." articles. --Tothwolf (talk) 00:50, 27 March 2011 (UTC)

A call for a "vote" is really a call for concise discussion to reach consensus about something. The following is how it's done. --Lexein (talk) 17:28, 27 March 2011 (UTC)

Inclusion criteria request for comment

This is not a vote. Briefly discuss your position (support, oppose, or neutral) "inclusion criteria" and what, if any, they should be. Wherever possible, provide wikilinks to appropriate sections of policy/guideline/essay. Should the criteria be No criteria, or a standalone article is required, or one or more third-party source(s), or a primary source which has details about the item, or a non-detailed primary source which has no details about the item? --Lexein (talk) 17:28, 27 March 2011 (UTC)

Lexein, This is not exactly what was discussed above. Your wording is decidely non-neutral and appears to be an attempt to sway discussion in the direction you want it to go. You still have not explained how removing information about historical software programs and information about programs readers wish to read about helps improve Wikipedia. I brought up reliable primary sources for simple facts about the software itself (features, options, version, operating system requirements, etc), which is per WP:SELFPUB. --Tothwolf (talk) 15:01, 28 March 2011 (UTC)

Noted, denied. My summary of positions in the RFC is carefully neutral, because it includes the full range of positions, including your apparent position that a mute source can be cited as if it actually said something (Orion's author). Though you write "reliable primary source", your definition of "reliable" is highly suspect (see Orion). You need to see that that author wrote nothing about the client, and none of the details about the client come from any sort of reliable source. As for how deletions help Wikipedia? That's not a proper complaint about the RFC, so I've answered in a new section. --Lexein (talk) 16:19, 30 March 2011 (UTC)

Discussion

• Support, one 3rd-party WP:RS is enough, per WP:TRUTH. Verifiability, for WP, requires 3rd party independent reliable source(s). Items here might not have enough notability for a standalone article, but still have enough verifiability for inclusion here. Primary source(s) are acceptable for details, as long as some of those details are supported in 3rd party source(s). --Lexein (talk) 17:28, 27 March 2011 (UTC)
• Support, one 3rd-party WP:RS per Lexein. (my dream would be that every feature is sourced...) mabdul 17:59, 27 March 2011 (UTC)
• Support, one article, or multiple 3rd-party RS for entries with no articles, to avoid the inclusion of entries that only have one review in one software website. --Enric Naval (talk) 19:50, 27 March 2011 (UTC)

• Comment: does this mean that one mentionsubstantial description or review in a book is insufficient? --Lexein (talk) 01:49, 28 March 2011 (UTC)

• In principle, no, it wouldn't be included. But that would depend on the book and the claims made in it, and it ought to be dealt with case-by-case. If the book is called "Extensive list of all IRC software under the Sun" then, no, it wouldn't be enough. If it's under a chapter called "Significant software that actually made a difference" or "Very popular software", or if the RS says that the program had significant influence, then the people on the talk page might reach a consensus that it's enough relevance. But in principle it wouldn't be enough, due to all those books with indiscriminate lists of software. --Enric Naval (talk) 15:27, 28 March 2011 (UTC)

I realize "mention" was too narrow. "Substantial description/review" is more of what I meant. Struck mention, above. And I agree that lists shouldn't be sources.--Lexein (talk) 22:31, 28 March 2011 (UTC)

Note: I can resign myself to a criteria of only one RS. I'm just hoping that at least it's a good RS. --Enric Naval (talk) 11:47, 13 April 2011 (UTC)

• Support, one article, or multiple 3rd-party RS for entries with no articles (for the same reasoning as Enric Naval). Edit: I should note that I also think all claims should be verifiable. IRWolfie- (talk) 21:25, 27 March 2011 (UTC)
• Require WP:V The notability guideline does not apply to article content. Reliable primary sources are acceptable for non-controversial facts per WP:SELFPUB.

  Simple facts such as a software's features, options, version, operating system requirements, etc do not require a 3rd party source. While 3rd party sources are generally preferable, 3rd party sources (especially books) lag far behind when discussing computer software because it is always evolving and being improved. Because of this, a software program's own documentation or source code is generally much more authoritative when discussing the software's features and functionality.

  Requiring 3rd party sources for simple facts ends up limiting this article's content to solely "Wikipedia-notable" items. This means we can't even include the first IRC client ever written, which was originally distributed with the early IRCd software for which we only have primary sources (and perhaps a few usenet discussions which mention using "irc software"). Per WP:N the notability guideline does not apply to article content, it only helps determine if a particular topic should have a standalone article. --Tothwolf (talk) 15:46, 28 March 2011 (UTC)

• Your argument is incorrect, because IRCd does have dead tree sources available from reliable publishers Integrating Linux and Windows Prentice Hall 2001, IRC hacks O'Reilly 2004, and from publishers that could be reliable or not Securing IM and P2P applications for the enterprise 2005, Ubuntu 10.10 Server Guide 2010 found via google books search [12]. The New Internet Navigator, Wiley 1995 and probably also The dictionary of multimedia: terms & acronyms Fitzroy Dearborn Publishers 1997 with another google books search[13] .

Notable IRC clients have RS books making non-trivial mentions of them. So, we should be requiring such RS before accepting a given client as notable. --Enric Naval (talk) 18:52, 28 March 2011 (UTC)

• My "argument" is quite correct, yet again you prove you really don't know this material... I was not talking about that IRCd. The IRCd I'm talking about is the first IRCd software cira 1988, and there are no dead tree sources or other 3rd party sources which discuss the IRC client called "irc" which was included with it. Why do you think ircII is called irc-two? It was the follow up to the first client.

  Like "irc", a very large number of historical and influential IRC clients do not have Wikipedia articles (i.e. WSIRC, XiRCON and many others), nor do all of them really all need standalone articles. Simple features and functionality are often better detailed in this comparison article and left as a redirect (as many have been for years). It makes no sense to have dozens upon dozens of stubs which do little more than list the version number and features, which is why we've merged and redirected them here per WP:ATD "Articles that are short and unlikely to be expanded could be merged into larger articles or lists.", WP:PRESERVE and WP:BEFORE. Those guidelines and policies are quite explicit that such merging is perfectly acceptable and even preferable.

  As I stated earlier, there is no requirement that a subject have a standalone article (per the notability guideline) before it can also be discussed within another article. Having such a requirement applied to articles would indeed greatly harm Wikipedia as it would often lead to a chicken or the egg dilemma. --Tothwolf (talk) 21:10, 28 March 2011 (UTC)

• Your argument is still wrong because the first, second and fifth books that I linked are about the 1988 original IRCd not about the second one. --Enric Naval (talk) 22:27, 28 March 2011 (UTC)

• No, those books still contain no mention of the client program whatsoever. Per your own argument they need to discuss the "irc" client program, not the IRCd daemon program. I note just as before you still choose not to respond to the other issues I raised. --Tothwolf (talk) 23:37, 28 March 2011 (UTC)

• Maybe that's why the client doesn't have its own article? Because only the daemon is notable? --Enric Naval (talk) 23:44, 28 March 2011 (UTC)

• Well, let's see... The very first IRC client ever... It certainly seems important for readers to know about... At the very least we need to cover it here in the comparison with other IRC clients... Secondary sources about IRC from 1988 searchable with Google? Not so much. Primary sources which detail the functionality of the software program? Quite difficult to locate but I found them. Perhaps the issue is notability does not nor was it ever intended to apply to an article's content, only to the subject of the article itself, which in this case is the subject of "IRC clients". As before, Enric, I don't think you or I have anything further to discuss. --Tothwolf (talk) 01:33, 29 March 2011 (UTC)

• No, maybe the issue is that it didn't have any relevant feature, and it didn't influence later clients. Or maybe it was a later client that made IRC really popular. But we can't know that without RS. No, we shouldn't be adding any entries that don't have any RS showing relevance. --Enric Naval (talk) 01:39, 29 March 2011 (UTC)

• Enric, we aren't making any such speculative or subjective claims in this article. We are noting factual data about the actual clients for which the clients' own documentation is perfectly acceptable per WP:V and WP:SELFPUB. Your radical notion of "notability" yet again noted (the notability guideline still doesn't limit article content), your logic is quite simply absurd. As far as importance to readers goes, if the first IRC client hadn't existed, how would later clients like ircII have been able to use it as the reference standard and be written with yet later clients using ircII as their reference standard? The first IRC client was written by Jarkko Oikarinen himself while creating the IRC RFC; RFC 1459 and as a reference standard software program, it was intended to be minimalist and did not include a lot of extra features. It was created well before DCC and even CTCP existed, both of which were first implemented in the ircII client. You cannot logically argue that the first IRC client software is unimportant to readers reading about this subtopic and should thus be excluded, although that is what you've systematically tried to do.

  Enric, we've already had this out with the WP:FICT debate. The notability guideline does not restrict article content nor does not prevent us from discussing what you personally deem to be "non-notable" items within larger articles (or for example lists of fictional characters). What it ultimately boils down to, is you would really like for the notability guideline to exclude and restrict article content, but thankfully it is explicit in that it does not. I'll note for newcomers to this discussion that you tried these arguments previously, even refusing to follow WP:BRD and edit war [14] [15] to remove the entry for KVIrc while making these same claims of "non-notability" and wikilawyering far more than anyone else I've ever come across. Those interested can read about it in the archives here.

  Enric, you still appear to only be here to somehow in your mind prove a point and only showed up here due to this disruptive AfD (background, additional links on request) where you argued "Delete all the non-notable entries." [16] I'll also note for newcomers that you were in the minority opinion, with the article being overwhelming kept intact. In any case, Enric, as far as this discussion goes, you and I are done. --Tothwolf (talk) 22:02, 29 March 2011 (UTC)

• So, the first IRCd does not appear in RS because it was not notable. And the really notable client was the second version ircii. And that second version does have RS mentioning it, just as one could expect from a notable software program. Sooooo, a strict inclusion criteria would work well in this case. --Enric Naval (talk) 22:52, 29 March 2011 (UTC)

• All, please keep other disputes (and "notes for newcomers") out of this discussion about improving this article. If you have a conflict with an editor, take it to your talk page, or their talk page, or please pursue WP:Conflict resolution elsewhere. As you find relevant reliable sources, please add them to the material in the article posthaste. If you have the energy to argue here, you probably have the energy to reliably source the table content where sources are lacking. Thank you. --Lexein (talk) 16:19, 30 March 2011 (UTC)

Enric Naval: "didn't influence later clients"? The very first IRC client, which was probably instrumental in the framing and development of the IRC specification and protocol? Do you realise how completely idiotic this sounds? By definition, the very first client and server for any (client-server architecture) network protocol have an extremely strong influence on later clients and servers, because without the first, typically experimental, implementations the protocol would simply not exist. I am forced to conclude that you not only have no experience in developing network software or protocol design, but further that you have never read anything by anyone who has, nor even devoted any thought to the question of how network protocols come into being in the first place. Incidentally, if you read chapter 8 (Current implementations) of RFC1459, you will find that the specification was published when the original ircd was at version 2.8 (and there were no other IRC servers at that time); trivially then the development of a server and client had been ongoing for some time before the specification was formalised sufficiently for other clients to be written to it, and thus the first client had a massive influence. You may also notice that that chapter does not make any mention of the first client; since early IRC clients were typically much simpler than servers (it being technically possible to use telnet as an IRC client if one knows the protocol well enough), very few sources discuss them. On a tangentially related note, do IETF RFCs count as WP:RSes? — Preceding unsigned comment added by Phase Theory (talk • contribs) 15:24, 24 April 2011 (UTC)

Again, I should follow more constantly, so I will comment only short: IETF standards are reliable, if the paper is a standard (not unknown useless drafts, exceptions are widely accepted "drafts" like the data uri scheme or IETFs April fools standard.) mabdul 16:08, 24 April 2011 (UTC)

Yes, IETF documents such as RFCs are considered to be reliable sources. Drafts too can be reliable sources (and I'm not familiar with any "unknown useless drafts", as Mabdul puts it). We cite a number of IETF drafts in the Internet Relay Chat article as well as many other Wikipedia articles. I found this had been discussed repeatedly while I was researching why we didn't have a citation template for IETF documents prior to designing and coding the {{Cite IETF}} template (which due to the complex nature of these documents, is probably one of our most complex but still easily used citation templates). --Tothwolf (talk) 20:10, 24 April 2011 (UTC)

• Support, one 3rd-party WP:RS as Lexein specifies. Most notably I would encourage keeping items that are not notable enough for a standalone article, but still got third party sources for enough verifiability for inclusion which also provide a minimum notability (while that one wouldn't meet the standards for a standalone article). If I'm actually too late to enter this vote, please remove my vote again (or add a fitting remark to it).Yarcanox (talk) 10:45, 22 April 2011 (UTC)
  • Again, how does doing this help us improve coverage of this subject? Such a proposal will result in excluding coverage of historically important subtopics such as the first IRC client (originally written by Jarkko Oikarinen) which was bundled with irc2.4. There are no "3rd party sources" (aka published books) which discuss the first IRC client at all, and it is not unique with this problem. There just aren't very many books which have given much more than a mention to most IRC clients, although there is at least one book which compares a few of the Windows and MacOS clients in feature comparison charts.

    Per WP:NNC we do not use the notability guideline to limit coverage of a subject (aka "require a 3rd party source for inclusion") within an article. While this article is technically not a list, per this RFC and WP:LISTN we similarly do not use the notability guideline to limit coverage of a subject within a complex list. What the RFC boils down to, is because the subject of "IRC clients" already meets the notability guideline, we can also create a "List of IRC clients". --Tothwolf (talk) 11:43, 22 April 2011 (UTC)

    • Here is one such comparison: Piccard, Paul L.; Baskin, Brian; Edwards, Craig; Spillman, George (2005-05-01). "Common IRC Clients by OS". Securing IM and P2P Applications for the Enterprise (1st ed.). Rockland, Massachusetts: Syngress Publishing. pp. 432–433. ISBN 1-59749-017-2. Table 16.1 IRC Clients and Features {{cite book}}: Unknown parameter |editorn-first= ignored (help); Unknown parameter |editorn-last= ignored (help) --Tothwolf (talk) 11:56, 22 April 2011 (UTC)
• Require WP:V per Tothwolf, plus a source (SELFPUB acceptable) indicating the existence of independent users. His arguments are, quite simply, correct; the other arguments made in this sprawling discussion are largely disingenous and/or sophistic. If we are going to consistently apply policy, then we ought really to list every client in existence. However, as noted, that is impractical since there are a large number (which cannot practically be estimated, and attempts to do so elsewhere on this page are flawed) of "toy project" IRC clients with no users besides the author (sometimes even no users including the author). Thus, I think the criterion for inclusion in this article should be any source indicating that there exist users of the client who are not authors of it. The source need not be a WP:RS, per Self-published sources as sources on themselves. Note that this is distinct from Tothwolf's SELFPUB argument, since I am using it not for the article content but for the notability test. Note also that it is possible for a WP:RS per Lexein's proposal not to constitute evidence of independent users, but typically only if there really are no independent users, in which case it's unlikely that such a RS would exist in the first place. As for the discussion of Verifiability, I think a combination of WP:SELFPUB and WP:BLUE allows us to perform the trivial OR that "yes, this client supports this feature" if it has documentation (which is not obviously inaccurate) or a changelog/release notes (again, excluding obvious fiction), and in addition can be tested to ascertain whether it exhibits signs of having the feature (eg. if a client sends, receives and displays CTCP ACTIONs correctly, and claims in its documentation that it 'has CTCP support', we would accept its claim unless we had evidence to the contrary). If it is not practical to test a client (eg. it is a binary-only client for a very obscure machine, like a Setun-70 or something - though actually that would be ternary-only, hehe - for which emulation is not available) then we would require 3rd party RS. If source is available, but the target platform is not, then drawing conclusions from source code is not really OR; it is WP:BLUE to deduce from, say, the existence of a non-stub function "handle_ctcp", that CTCP is at least partially supported. Conflict disclosure: as noted below I am an author of an IRC client, which may be a conflict of interest; on the other hand, it does indicate that I am knowledgeable on the subject in general (and I would like to mention that essentially all technical points made by Tothwolf are accurate). PT (talk) 15:06, 24 April 2011 (UTC)

• Require third party WP:RS and the source should be significant. SchmuckyTheCat (talk)

In re deletion of original research

In General

How does deletion of original research help Wikipedia and this article? That's easy. WP:OR harms Wikipedia's credibility, because it elevates unsourced observations by editors to the level of reliably sourced material. "Wikipedia says" becomes worthless if the sources aren't independent, published, reliable, and verifiable. Wikipedia's responsibility is not to the notion of, as Tothwolf describes, what "readers wish to read about", its responsibility is to the sourced facts: to present facts already reported by others, and not "simple facts" as observed by editors; see WP:TRUTH. Wikipedia, and all formal encyclopedias, are not about the thing itself, they are about reportage about the thing by others.

More specific

As a side note, this is the reason why source code and binaries are poor WP sources - they are the thing itself (in alternate forms), not "writing about the thing." It's OR to describe pottery color, shape, weight, glyphs or country of origin without reliable sources. It's OR to dismantle a toaster to describe its parts without reliable sources. In that light, it's OR to describe software details or features which have not been discussed in independent sources (and OR to describe based on source code or binaries).

About IRC clients

I don't oppose software documentation as a primary source, since it is descriptive prose, but I'm of the opinion that it's insufficient for inclusion, without discussion in at least one 3rd-party source. And in the case of no such documentation and no 3rd-party source and no discussion even in newsgroups (for old clients), sorry, no inclusion. --Lexein (talk) 16:19, 30 March 2011 (UTC)

• That's all fine and good, however noting simple facts about a software program using the program's own documentation per WP:SELFPUB or even source code or documentation embedded within the source code when we have an open source program (per WP:RSUE) does not constitute original research. Original research would require us to draw our own conclusions or make subjective statements such as "x is better than y" based on such simple facts, which we do not do within this article. Per WP:SELFPUB it is acceptable to note such non-controversial facts "software x supports Microsoft Windows and requires plugin z" using a primary source, which as I noted previously, is going to be more reliable than a book because of the delay in the publishing process of a book compared to the rapid and constant evolution of software.

  As for binaries, binaries are not source code and in general we can't cite them. I've not previously brought up binaries and your mention of them here appears to be done solely a means of discrediting me and drawing focus away from the main discussion. About the only thing we could possibly cite with a binary is "program x's binary is y kilobytes in size" which would be per WP:BLUE. Other than that, we can't use them for much as binaries cannot be read as plain text by a human. Source code on the other hand is plain text, and while not entirely in English, it is no different from using a source written in Russian, Polish, or any other non-English language. Verifying the source will obviously require someone to be somewhat familiar with that language (at least the structure and main components) which is why per WP:RSUE we generally prefer English sources when they are available.

  As previously noted, your own idea of mandating a 3rd party source before we can discuss something in this article ends up going too far in the other direction. This idea ends up violating WP:NPOV by presenting the material within the article in a skewed Wikipedia-notable only way and eliminates coverage of material important to the parent topic of "IRC clients", including coverage of the very first IRC client and other historical or influential clients which predate Google by nearly a decade. This is again why the notability guideline does not limit the content of articles.

  Lexein, now, if there was actual money to be made spamming IRC clients on Wikipedia and had this ever actually been a problem here, we would have already taken steps to prevent it. This however has not been a problem, and your proposal is a solution in search of a problem. --Tothwolf (talk) 23:14, 30 March 2011 (UTC)

You've got a critical error in your discussion of WP:OR policy. You may wish to strikethrough some of what you wrote: Quoting directly from policy WP:OR:

"Wikipedia does not publish original research. The term "original research" refers to material—such as facts, allegations, ideas, and stories—not already published by reliable sources. It also refers to any analysis or synthesis of published material that serves to advance a position not advanced by the sources."

The one-off author of Orion cannot fairly be termed a reliable Wikipedia source; we agree on that. Even if he was, he didn't publish any separate documentation about the program, we agree on that. So your invocation of WP:SELFPUB doesn't apply because the author wrote nothing about Orion. So you invoke the source code as a source (as WP:SELFPUB?) about itself, but that can't work because the source code isn't an author, it's the object-under-discussion itself; as a result, there's no reliable source at all. The only way to verify that the source code meets its claims is to run the program, which is further original research and/or compile the source: even further OR. At Wikipedia, we have agreed not to be evaluators of WP:TRUTH, or publishers of self-observed "truth" - we have, as a group, agreed that only verifiability matters, as established by external, independent, reliable sources. WP is WP:NOT Consumer Reports or a research lab, or a software documentation service. As for Orion, we're back to original research, and we're done, and Orion's out, in my considered opinion. Do you still assert that Orion merits inclusion, even though its details are zero-reliably-sourced? Yes or no, at least.

The now quite old and well-established Wikipedia-wide bedrock requirement for reliable sources (not my own idea, by the way, but that of thousands of Wikipedians) stands in the way of including Orion. There are lots of clients with lots of sources. Let the clients without reliable sources go, move them to Talk until a source is found, and give yourself a little peace. If they were really all that important, wouldn't have somebody written about them? As for encyclopedias having bias toward published sources, you're absolutely right. It's not undue, though, because that's how encyclopedias strive for accountability and reliability; it's the only way.

BTW I'm only proposing one independent reliable source. Other editors seem to want articles or multiple sources, so your battle isn't really with me. --Lexein (talk) 08:51, 1 April 2011 (UTC)

Lexein, I suggest you drop it and move on. Filing an RFC to try to gain an advantage because you don't like the way long standing Policy is written and that I can cite it when you began the wikilawyering is inappropriate.

Consensus at AfD (latest such example) and on other comparison article talk pages is that these articles are encyclopedic and do not violate WP:NOT. There is absolutely nothing wrong with giving coverage to subtopics (be it "red links" or "non links") which don't qualify for their own standalone article under the notability guideline. Your attempt to mandate a 3rd party source as inclusion criteria is your own vain attempt at trying to force the notability guideline to restrict article content.

Moreover, you continue to gloss over (and ignore when challenged) the fact that you are also trying to exclude the very first IRC client ever written, which came bundled with the original IRCd.

Good grief, we have a List of recurring characters in The Simpsons (and I would suggest not trying to argue against using The Simpsons episodes themselves as self-published sources as others have already tried and died). Not one of those characters is considered "notable" on its own, which is why we tend to merge and redirect such content and simple facts into larger lists and articles per the Alternatives to deletion section of the deletion policy which states: "Articles that are short and unlikely to be expanded could be merged into larger articles or lists.", and per the WP:PRESERVE section of the editing policy and WP:BEFORE section of articles for deletion.

It is far better to merge and redirect articles like the one we had for Orion to a larger article as it makes little sense to have a stub article which is little more than a feature list and version number.

Lexein, When your earlier tactics and arguments here to exclude what you personally consider "non-notable" content failed, you began this latest tactic of trying to put words in my mouth and twisting anything you could in order to try to discredit me. No one is falling for your straw man arguments.

You showed up here out of the blue, having never edited this article, nor contributed to articles within the parent Internet Relay Chat topic, and just like Enric Naval began pushing to remove all "non-notable" content and coverage of subtopics (in your mind) from this article. Unfortunately for you and Enric (who oddly enough showed up here in a similar way), the notability guideline does not apply to article content. Full stop. You and Enric both have wikilawyered this to death and continuing to try to claim it so won't change that.

Lexein, given your "tactics" and your apparent motivation, [17] [18] I don't see a point in continuing this discussion, so I'm done discussing this with you. I again suggest you move on and find something else to do which doesn't involve POV-pushing your own view of the notability guideline. --Tothwolf (talk) 04:50, 2 April 2011 (UTC)

You're wrong on policy, so you've shifted to insinuation, false accusation, and personal attacks. My only agenda is demanding actual reliable sources(of any kind), which you should respect as written in policy; your attempt to paint that in a bad light is disrespectful to the WP:Five pillars foundation of Wikipedia. I stand by everything I've written above, which counters your every attempt to rationalize unsourced content, content sourced by itself, and original research, focusing on Orion.

It's important to the improvement of all articles that unsourced material be removed, and that editors not pretend that unreliable sources, or nonsources, are reliable. I put no words in your mouth - I do reasonably expect that you agree that a silent source is not a reliable source, and that objects are not sources, exceptions for broadcast media notwithstanding. As pertains to this article and Wikipedia in general, I stand by this: it is a disservice to readers, and always will be, to present unsourced material as if it were reliably sourced.

I posted the RfC (others requested it, if you read back) only to assess the state of consensus, which illustrates that most editors here want at least one independent reliable source for claims made, which is precisely on policy, though it's not what you want.

As for your insinuations, I have nothing to do with Enric, and as I explained in my private email response to you, I welcome any challenge you may wish to put to my edit history and standing here. Bring it on. You are trying to keep unsourced material in this Wikipedia article, which is in blatant opposition to a bedrock policy, and trying to keep it there by a variety of edge-case reasoning. This is a slippery slope which leads to an unsourced Wikipedia, which tens of thousands of editors do not want. WP:OTHERCRAPEXISTS, such as The Simpsons, but is not relevant here.

You've researched my edit history: good. You saw that during an AfD, I and two other editors proposed clear inclusion criteria and advocated a search for better sources, and I'm proud of that work. My motivations are above reproach. You should have noticed that I did edit ChatZilla and others to add reliably sourced content a while back, so gee, I guess I didn't show "up here out of the blue, having never edited this article, nor contributed to articles within the parent Internet Relay Chat topic". Ouch!

You falsely accuse me of trying to "exclude the very first IRC client ever written, which came bundled with the original IRCd"; I have made no such effort. I suspect that sources for it exist which have not yet been found, such as dead-tree books, or even newsgroup items by notable authors (the old ones) and am willing to assume good faith about that. But not Orion, which merits no inclusion, lacking every kind of reliable source. There is a difference there, if you'd stop for a moment and think about it.

Your preferences aside, it is in every article's and every editor's best interest, to push for actual policy-deemed reliable sources for every challenged claim in an article, and not to pretend that blatantly unreliable and unusable sources are reliable. Usually, it's made easy by well-written comprehensive sources discussing the item under discussion. Here, sadly, there exist no such sources for Orion, not even documentation. So, I guess we're done, except for dispute resolution. Keeping out unsourced material (or properly sourcing it) is neither a dead horse, a red herring, nor a straw man: it's policy. Too bad you don't see it that way. --Lexein (talk) 16:19, 2 April 2011 (UTC)

What you mean is, since excluding irc is clearly wrong, you choose to ignore the fact that your proposed rule would exclude it, and make up some guff about what you suspect. You also cite AGF; I cannot parse your meaning here, as sources can't act in bad faith. You are trying to push a rule which clearly does not work, and refusing to accept that it does not work. You are actually making it very hard for others to AGF, but I will try. Your positions are not supported by policy, nor logic, so put down the stick. Please. PT (talk) 16:33, 24 April 2011 (UTC)

You should have stopped at "I cannot parse" and left a question. My position about assuming good faith is supported in policy; as an editor I have discretion to assume good faith about the existence of sources (this is why the tag {{Citation needed}} exists), your pointed misrepresentations notwithstanding. I trust that sources (dead-tree, RFC, newsgroups) do exist for irc because I AGF about Tothwolf's claims about irc: it's old, it is sure to have been discussed by author and users, and therefore is quite likely to be supported in as yet undiscovered reliable source (independent or primary). I'm willing to allow six months to find any such source; I've been looking, too. However, I cannot extend the same presumption of existence of sources to Orion: it is, by comparison, new; its existence has been purely online, yet there is no evidence of it ever being discussed by anyone, anywhere, not even its author, in any verifiable source (not even blogs, forums, or other wikis). If I'm "pushing" (your word), it's against the inclusion of material based on unreliable sources (without "the author describes" distancing language), silent sources, purely primary sources unsupported by any other sources of any kind, or original research. I trust you understand my meaning now. If not, I refer you to Keeping out unsourced material (or properly sourcing it) is neither a dead horse, a red herring, nor a straw man: it's policy. Too bad you don't see it that way. --Lexein (talk) 09:45, 28 April 2011 (UTC)

I think this would be an improvement if we also allowed big third party sites. E.g. if you take the deletion of InspIRCd happened mainly due to lack of print sources, although SearchIRC lists it as second to most used IRCd - and I think deleting such a software makes the whole Wikipedia IRC section worthless.

IRC isn't an area with lots of print sources and if we totally stick to them, we could aswell also just leave the work in this section altogether (which is what I partially engaged in the last months). But I agree that a bit of cleanup in the IRC clients listing would be appropriate, if it doesn't lead to a list of 5 clients - which is what I am really afraid of taking the massive deletions on Wikiproject IRC the last months into account.

So I suggest adding a weak policy for inclusion first (e.g. one more or less large 3rd party site mentioning the client) and if, after removing all the clutter based on that, the list is still full of irrelevant information and clients noone uses, establishing a new discussion about a more strict policy. But please don't rush deleting 90% of the article including the interesting parts, that has happened way too often. Yarcanox (talk) 18:15, 14 April 2011 (UTC)

Let me add a separate comment not on inclusion criteria but on original research. I think having information with no sources is bad, but IMHO for purely technical information (that is, no studies on impact or userbase or awesomeness) the client's original documentation is fine and should be accepted. So we should attempt to put such references in where they are missing (practically everywhere I suppose) and if that leaves us with totally original research, come back to discuss if we can rid of it in an appropriate manner. (Yes I'm a friend of taking it slow... I hope we can get this article in small steps towards a clean state rather than just removing half of it in one day) Yarcanox (talk) 18:18, 14 April 2011 (UTC)

Criteria

It seems that the majority are in favour of at -least- mention of the client in one reliable third party source, should this then not be the grounds for inclusion? IRWolfie- (talk) 18:44, 11 April 2011 (UTC)

WP:V is policy and WP:N does not dictate or limit the inclusion of material. So no, the majority are not in favour of this. The "majority" is not made up of a couple of editors who've never contributed to this article (I've been contributing to it and the related articles heavily for years). If this ends up going to AN/I, given the past targeting of this and other articles I've contributed to, including the AfDs followed by blanking with sockpuppets, there will be some serious questions raised about some of the editors who showed up here. --Tothwolf (talk) 19:50, 11 April 2011 (UTC)

We want to contribute (and yes that means removing clients which don't seem to have grounds for inclusion) but that requires firm criteria and the majority of those who commented are in favour of requiring reliable third party sources discussing the client as grounds for inclusion or an existing article. The discussion has no connection to your experiences in other articles and I have seen no mass blanking: the criteria will help improve this article not make it worse. I should point out there is an existing template for stand-alone lists: Template:Stand-alone_list "This is a stand-alone list. Please only add subjects that have a Wikipedia article." also see Wikipedia:Stand-alone_list#Common_selection_criteria. A comparison article and a list are effectively the same thing; in fact this article appears to be a combination of lists (first two tables) and comparisons. IRWolfie- (talk) 20:43, 11 April 2011 (UTC)

Replied [19] in the section below. --Tothwolf (talk) 22:04, 11 April 2011 (UTC)

...and replied again here. --Tothwolf (talk) 14:30, 12 April 2011 (UTC)

I should point out if there is local consensus to require a source or a certain type of source, then that's OK. Just don't call it notability, because it's not notability. Gigs (talk) 13:10, 22 April 2011 (UTC)