Talk:Comparison of IRC clients/Archive 4

The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

This is an archive of past discussions about Comparison of IRC clients. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 1

Why all clients aren't listed

I'm just going to ask a simple, blunt question. I'm just a simple guy that was looking for IRC clients, and was dissapointed and just how small this comparison list was. I went to the talk page... and found a multi-page argument about 'Firm Rules for Inclusion' and bickering over Orion. I'm gonna be honest: Lexein, IRWolfie, why exactly are you trying to make this into the equivalent of a Favorites list for IRC clients? I'm not a huge wiki editor, I can't link rules left and right, but I know that being a dick is looked down upon, and I have to call you both out on it. This is a comparison list of IRC clients. That means it compares IRC clients. Not popular IRC clients. Not Well Known IRC clients. Not your favorite IRC client. All IRC clients. Anyone who attempts to argue this otherwise promptly gets very displeased, nearly rude response rebuffing them from you both. So what's up, you two? What exactly do you have against this page? 75.218.251.9 (talk) 06:04, 27 June 2011 (UTC)

All IRC clients isn't the purpose of the page, or Wikipedia as a project. SchmuckyTheCat (talk)

What's up? I agree with SchmuckyTheCat's reply. We're not going for a "Favorites" list, nor claiming to be all-inclusive, nor attempting to be rude. For me, it's about reliability. Wikipedia aims to be an encyclopedia, not a catch-all; read WP:Five pillars.

Ideally, all IRC clients would be cataloged and compared en masse by other reliable sources. But they're not. This list, unfortunately, has had to be built up manually from individual sources and existing Wikipedia articles (which exist due to sufficient reliable sources). I'm in favor of all clients listed in this list/table being supported by at least one independent reliable source. Otherwise, it's just a list of random self-proclaimed clients with no independent verification. I'm even willing to go so far as to try to establish reliability for sources which at first appear not to be reliable.

The "bickering about Orion" was not (for me) about Orion - it was about verifiability and original research. Luckily, there's no WP:DEADLINE.

The point: the voice of Wikipedia is the voice of the reliable sources which we cite. If no reliable sources speak on a topic, then neither does WP. Hence, there are few IRC entries in this comparison table. See WP:NOTTRUTH and the surprising FLAT. --Lexein (talk) 19:36, 27 June 2011 (UTC)

I actually have an offline draft where I've gone back through the older revisions to pull salvageable content and reintegrate it with the newer templates. I need to make one or two more passes over it and make a few more template changes, then we can begin to add sources for more of these entries. Unfortunately, as I discussed with Lexein, due to template changes and improvements to other sections of this article, it wasn't possible to simply revert back to a previous version and update from there. I'd planned to have it done by now, but I'd greatly underestimated just how time consuming the task would be. --Tothwolf (talk) 22:16, 27 June 2011 (UTC)

Looking forward to it! This stuff does take a lot of time, that's for sure. I just hope you get it to a usable state before you completely burn out or lose your mind. I sorta wish Cirt would take a stab at some of the harder to source items - he has access to a university library, apparently, and sources I'd have to pay hundreds of dollars to access. --Lexein (talk) 00:30, 29 June 2011 (UTC)

Inclusion of Specific Clients

These are the specific clients I have issue with being included LeetIRC, nexIRC, Neebly, KoolChat, DMDirc and AmIRC and ii. I admit that I don't know much of anything about these clients, but it is for those who wish to have them included show why so, not for me to justify their exclusion. Is there any current justification for their inclusion. note: if a client is being included because you think it is interesting then you are implicitly applying notability as a criteria. I have no objection at this time with any of the other clients currently in the article. IRWolfie- (talk) 23:05, 21 April 2011 (UTC)

No, it is up to you to justify your actions here. If you admittedly don't know much about this subject, then go study up on the topic before proposing that we remove material from the article. AmIRC even meets the notability guideline, which is why I've kept it a red-link while I have a draft in my userspace.

Further, what you are pushing here still amounts to instruction creep because as has been brought up above, and can be seen in the edit history of the article, we've never had much of a problem with "spammers" here. There just isn't much money to be made in "spamming IRC clients", and especially the Freeware and Open Source clients you are mentioning now.

IRWolfie-, care to have a go at the question I posed above for Enric since he apparently has no intention of trying to answer it? --Tothwolf (talk) 08:23, 22 April 2011 (UTC)

I did not say I don't know much about the subject, I know of many (heard of, not used, I know use irssi exclusively) of the clients listed. I said I don't know much about these specific clients since I have -never- heard of them. I think there is no reason to justify their inclusion that I have seen, but maybe I can be swayed with the right evidence. Also I have no interest in answering this colour question you refer to: since I would be about 6 to 8 at the time, also see Wikipedia:Expert_editors Nor does Wikipedia grant users privileges or respect based on subject-matter expertise. IRWolfie- (talk) 08:52, 22 April 2011 (UTC)

So, if you are familiar with irssi, where exactly would you classify BitchX or EPIC (notice someone removed this one from this article)? I'm sure you've heard of those at least. How about ircII? You do realize that the person who first began removing material here initially tried to have all of these deleted as non-notable? If you are going to quote WP:EXPERT, then I'll do some quoting of my own. Let's start with WP:NOTPAPER: "Wikipedia is not a paper encyclopedia, but a digital encyclopedia project. Other than verifiability and the other points presented on this page, there is no practical limit to the number of topics Wikipedia can cover, or the total amount of content. [...]"

...and how about some quotes from Wikipedia:Five pillars:
"Wikipedia is an online encyclopedia. It incorporates elements of general and specialized encyclopedias, almanacs, and gazetteers. [...]"
"Wikipedia has a neutral point of view. We strive for articles that advocate no single point of view. Sometimes this requires representing multiple points of view, presenting each point of view accurately and in context, and not presenting any point of view as "the truth" or "the best view". [...]"
"Wikipedia does not have firm rules. Rules on Wikipedia are not carved in stone, and the spirit of the rule trumps the letter of the rule. [...]" --Tothwolf (talk) 10:22, 22 April 2011 (UTC)

BitchX is fine for the article since it has it's own article. I don't see why EPIC is not included since third party references can show its notability. Also since wikipedia does not have firm rules that does allow for applying an inclusion criteria to the article. IRWolfie- (talk) 21:46, 25 April 2011 (UTC)

EPIC was removed along with numerous other clients such as PJIRC and others when a large portion of this article was blanked following the mass-AfD disruption.

When people discuss the topic of text based *nix IRC clients, the main ones that come up are: ircII, Irssi, EPIC, ScrollZ, and BitchX. All of these have had a major following and tons of users, but again due to the dynamic nature of the web and internet as a whole (including stuff which used to be readily available via FTP), it isn't as easy to find information on some of these as it once was.

"Also since wikipedia does not have firm rules that does allow for applying an inclusion criteria to the article." ...I think you have that backwards, as I said before:

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" for this subject.

What this means is the whole argument of "require a 3rd party source for inclusion so we can exclude "non-notable" IRC clients" isn't going to fly. You can't prevent those of us who wish to improve Wikipedia's coverage of IRC from discussing any of these clients in Internet Relay Chat or an eventual Internet Relay Chat clients spinout article, or in Comparison of Internet Relay Chat clients so long as the material itself is verifiable, which can also include using the documentation and source code of a program for verification of its features and functionality. --Tothwolf (talk) 03:47, 26 April 2011 (UTC)

"and source code" Bzzt. Although documentation about the subject may be cited as a primary source, because it is descriptive prose, written about the subject, by the author, source code should not be used as a Wikipedia primary source without distancing language per WP:RS("the source code lists the following features") because a) it's a transformation of the subject (it's like a WP editor citing the text on a vase - it's obvious Original research) and b) it's unreliable without verification(ask any senior software engineer): verification of source code requires original research (inspection, compilation, functional testing), which is not usable in Wikipedia articles. Wikipedia's voice should only be used to speak the words of reliable (independent or primary) sources, never unreliable ones without explicit distancing language, and never original research. --Lexein (talk) 02:45, 29 April 2011 (UTC)

“

Instructions such as computer code, which are intended to be executable by a computer, will often convey information capable of comprehension and assessment by a human being. [1] A programmer reading a program learns information about instructing a computer, and might use this information to improve personal programming skills and perhaps the craft of programming. Moreover, programmers communicating ideas to one another almost inevitably communicate in code, much as musicians use notes. [2]

”

— Newman, Cabranes and Thompson, Universal v. Reimerdes (DeCSS)¹

1. ^ Universal v. Reimerdes, 273 F.3d 429 (2nd Cir. November 28, 2001).

Lexein, either provide a link which explicitly states source code is unacceptable per WP:V and WP:SELFPUB or drop it. It doesn't matter how loudly you exclaim otherwise, using source code for verification of a program's features is not original research. --Tothwolf (talk) 04:36, 29 April 2011 (UTC)

That quote does not support the position that some random unknown blighter's software source code constitutes a reliable, verifiable source on Wikipedia, nor the claim that no expertise is required to understand source code. The quote indeed invokes "human being", but only as differentiated from a computer, and only referring to programmers, which is~~, oh look,~~ the very next word in the quote. ~~Ouch.~~(joshing --lexein)
Consider: if source code were to be cited, which part? Whole releases? Code blocks? Statements? Function names? Comments? Documentation embeds? Readmes? What if the comment says, "connect to proxy" but the code does no such thing? Or if the code seems to, but waaaay down in the header files, that function call is #define ConnectToProxy (...) null macro'd? It's a Pandora's Box of original research. Per WP:Verifiability#Original_research, "Sources must support the material clearly and directly" - presumably, without internal conflict or vagueness. Would the file name and line number be cited? Hm. If not, why not? Interestingly, no such information seems to have been listed in citations so far. It's an interesting test case. I hereby double-dog dare you. ~~We'll just invite some serious deletionist anti-original-research crusaders I've had dealings with in the past to the shredding-of-the-tissue-of-bad-ideas party.~~(joshing--lexein)
As for your demand that I provide a link, I need not, nor need I drop it. Your attempt to invert responsibility is inappropriate, and goes against WP:CHALLENGE, WP:BURDEN, WP:RS, WP:OR, etc., etc. Instead, you must prove the acceptability of source code as a reliable source which can be used without support by any other source, and without appropriate distancing language, and do so without resorting to the non-applicable foreign language provision of RS, or you must prove that editors describing a subject is not original research. Also, if you wouldn't mind, please prove that source code has been used "since the beginning," or, as you have demanded so many times, "drop it." --Lexein (talk) 13:03, 29 April 2011 (UTC)

Lexein, you know, on one hand you tell people to stop making personal attacks, "WTF? Everybody, goddamnit, stop attacking editors." [3] yet on the other hand you turn around and do the very same thing while attempting to disguise the personal attacks as "policy arguments", which are primarily made up of nothing more than fallacies.

I realise you didn't like the fact I could cite Universal v. Reimerdes ~~and I fully expected you to wikilawyer when I did so~~. That does not change the fact that it states what it states. "Instructions such as computer code, which are intended to be executable by a computer, will often convey information capable of comprehension and assessment by a human being." Source code is a text language that can be read and understand by a human and is no different from any other non-English source. "Moreover, programmers communicating ideas to one another almost inevitably communicate in code, much as musicians use notes." As a programmer yourself, you know this is true, and while you understand that my arguments have merit, you have become so fixated on "winning" your argument with me that you can't see the forest for the trees and are willing to throw out everything in order to do so. To take this one step further, if we can cite sheet music for simple facts, we can similarly cite source code for simple facts. WP:NOTPAPER states: "Wikipedia is not a paper encyclopedia, but a digital encyclopedia project. Other than verifiability and the other points presented on this page, there is no practical limit to the number of topics Wikipedia can cover, or the total amount of content."

Lexein, this is the last time I'll state this. If all you want to do is fight and attack other editors such as myself who disagree with you, then I see no point in attempting to continue to discuss any of this with you. While you and I may not always agree, I will continue to improve content here on Wikipedia using the best available methods and sources at my disposal. If you are unwilling to at least consider my points, coming from someone who has invested 100s of hours into this article and even more into related computing articles, then for the benefit of Wikipedia and everyone else (anyone who has had the patience to fully read our back-and-forth arguments deserves a barnstar or two), it might be best if you and I go back to working in the subject areas we generally tended to work in prior to our argument and do our best to avoid crossing paths in the future. --Tothwolf (talk) 02:28, 30 April 2011 (UTC)

No. Do not assume bad faith about me, don't parse so narrowly, and don't see attacks where there are none. The overall meaning of the quote is that programmers are the intended human audience for source code, and possess skills and training suitable for its interpretation. At Wikipedia, we rely on sources which unambiguously make statements, which we then cite. Source code is not necessarily such a clear and unambiguous source for Wikipedia, due to the realities of computer language implementations, subtleties, and the fact that code can differ from comments which can differ from documentation which can differ from helpfiles. And, for open source code, which can vary considerably over time and may not have any fixed releases, just a repository, it's even worse. And don't make that out to be an attack, either. It's just a position which has merit.
Sheet music is reliably published by big companies with editorial policies and even a set of standardized formats. Big difference with the source code written by a one-off random blighter in a garage with a website. Sheet music date, author, publisher, key, tempo, are clearly stated with a common language without complicating internal ambiguities. This stands in magnificently stark contrast to, as stated previously, source code which might be internally consistent, but more often is simply doomed to be vast miasma of original research.
You've often told people to "back off" and "go away". That's not the Wikipedia way, and it's not my way. If you honestly felt attacked, above, sorry - there's joshing, and there are attacks, and I was joshing, at worst. I've ~~struckthrough~~ what I think might been misinterpreted. You could do the same, for some of your harsher rhetoric and accusations, by the way, nudge, nudge, wink, wink.
It's natural, when arguments feel personal when they really aren't, to shut down the discussion. My advice is to instead, WP:DISENGAGE for a while, as I sometimes do. It's helpful. Oh, and I really do mean read that thing.
I must, without rancor, stand by my remark about inversion of responsibility. I'm sorry, but your argument isn't with me. WP:BURDEN does rest with the adding or restoring editor. But I do believe in the entirety of it - I seek sources and AGF about offline sources.
I'd appreciate it if you'd answer my challenge about trying to actually _use_ source code as a source, and its pitfalls. There is value in that discussion, and it could escalate to the establishment of an essay or guideline on allowing, and under what circumstances, such usage. You should see that as long as there is no explicit criteria here, and no guideline about source code there, there will be endless, frequently fruitless debate, endlessly restarting. We're farther along that track now, and some actual concrete good could come from this and other discussions. I say, no point in giving up now. Though I do see definite value in getting back to articles, too, as I explained in my email.
Your interpretation of my attempting to focus on policy as another "personal attack" is uncalled for, as it keeps focus on editors, not articles and yes, policy. Again I invite you to actually read WP:TIGERS. You're not a victim here. I have no desire to see you stop editing, or stop discussing, here, or anywhere else, though I do wish you'd sometimes see other points of view. --Lexein (talk) 09:55, 30 April 2011 (UTC)

In some cases source code is very clear, as described further below with ircII.

As for articles with source code citations, one good place to start is Category:Articles with example C code, or better yet Category:Articles with example code. For something a little more challenging, try International Obfuscated C Code Contest.

Again, Wikipedia is not paper. The general structure of Wikipedia allows a reader to read to whatever level of detail they desire. Not all material on Wikipedia is going to be within the grasp of every single reader. For example, Category:Astrophysics, Category:Molecular biology, Category:Mathematical and theoretical biology, Category:Particle physics, Category:Theoretical physics, heck even Category:Organic chemistry.

Not all source code will be able to be understood by a general audience. The same holds true for non-English sources where the reader is a native English speaker. This doesn't mean we can't cite source code. If we had to cite sources which anyone could understand, most of the categories I linked above would be empty of non-existent.

On a side note, the best way to handle "unreleased" code in a repository is to link to a static revision id, much as we do when we have to link to the static version of a wiki page. --Tothwolf (talk) 14:49, 30 April 2011 (UTC)

The size or detail of WP is no issue. I do wish you'd taken my request seriously. This portion of the discussion is about citing source code as a WP source, especially from a primary source, to support claims about software features. The examples offered are not concrete examples of this, but are in fact, examples of the few acceptable uses of source code in Wikipedia, which are not in dispute here. Nearly all the articles are about the practice of software engineering; a few are about specific software products. But still, none of the articles offered use inline citations to primary source code files for features of a software product itself, or anything else. I examined dozens of the articles at random, I'll describe four representative examples here:

Source code examples in software articles such as POV-Ray and Verilog are about the language embedded within the product, which is reliably documented, for the purpose of illustration of the embedded language, not the software's implementation language. There is no quoting or inline citation of the software source code, or the embedded source, to support claims. (Also, the extant examples are OR, as they are not quoted from any source.)
gets doesn't inline cite or quote any real online or offline source code. Only man pages and standards documents are cited. (Also, the included example source is WP:OR - it's not from any source).
International Obfuscated C Code Contest is about the practice of software engineering, and the quoted code blocks, inline cited from the IOCCC (refs #10 & #11) (not the original author), are short, unambiguous, and complete; claims made about the features of the obfuscated software are not cited from its source code, but from the IOCCC.
Literate programming shows source code which is quoted from sections of published academic articles which include source code (ref #10). The article is cited, not source code files.

My position remains that quoting or citing of source code files from primary sources hasn't happened, not for software features, not where code is discussed for the purpose of illustration, nor where it is quoted in toto (an RS quoting it is used instead), or anywhere else. I believe it is, in fact, universally considered original research and/or too weak as a primary source. So, in my opinion, it shouldn't happen here. --Lexein (talk) 14:06, 1 May 2011 (UTC)

No, the argument has gone back and forth that "source code can't be understood by non-experts so we can't use it" and "source code isn't a reliable source, ever, and we can't use it". Neither of these arguments is true. The real argument from those making such arguments is that citing source code will lead to the end of the world.

Source code is reliable and usable per WP:SELFPUB when noting that a software program supports a feature. Source code will generally even be more accurate of a software program's own documentation included with the software (<sarcasm>don't you know, all software developers always keep their external documentation updated when they modify code!</sarcasm>) and a published book (due to the lengthy process of publishing and printing a book as outlined above).

As for the non-experts argument, some source code might not be readable by a layperson (not all), but that does not mean we can't cite it.

"ircII supports CTCP FINGER" –
<ref>http://www.example.com/</ref> – "ircII IRC-ohjelma tukee CTCP sormi"
<ref>See line 267 "do_finger()" of 'source/ctcp.c' included with ircii-2.1.4e.tar.gz</ref>

Both examples would require the person verifying the source to be able to read a non-English source. The second is going to be more reliable. --Tothwolf (talk) 15:14, 1 May 2011 (UTC)

From WP:PRIMARY "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. For example, an article about a novel may cite passages to describe the plot, but any interpretation needs a secondary source. Do not analyze, synthesize, interpret, or evaluate material found in a primary source yourself; instead, refer to reliable secondary sources that do so. Do not base articles entirely on primary sources." (emphasis in the original). --Enric Naval (talk) 07:30, 29 April 2011 (UTC)

Reading source code does not require "specialist knowledge". Source code 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.

"Program X includes feature A, B, and C." [cite: program's source code] is perfectly acceptable and if you don't like this then you are going to have to take your argument to a much larger venue because this has been acceptable practice here on Wikipedia since the beginning. I did note you dropped your "source code isn't a reliable source" argument, so I'm not surprised to see you trying to now claim "interpreting source code requires specialist knowledge" [4] (redacted or otherwise) even though there isn't any interpretation required. --Tothwolf (talk) 08:15, 29 April 2011 (UTC)

Errrr, Tothwolf, I'm afraid that we are not talking about the same thing. I and Lexein are talking about citing the actual C (programming language) statements. A non-programmer will not be able to interpret them. Citing actual source code is not an accepted practice, see how Template:cite source code is a red link. In IrcII, for example, it's not citing the source itself, it's really citing the NOTES and UPDATES files, which explain the cited fact in plain English; the link to the actual code file could be removed but it's good to have it there for verification purposes. Significant comments inside the source code are probably citable too. Making your own analysis from an uncommented function in a code file wouldn't be OK.

However, they are still primary sources and we shouldn't have entries in the list that are based only around them (in the same way that we shouldn't have an article that is only based in primary sources). Once a client has an entry in the list (because it has third-party RS) you should be able to cite specific features from plain English comments in the source code and the project files. These texts can be taken as being part of the documentation of the project, as a last resource for features that don't appear in the documentation manuals because they are too specialized. I think that this has already been told to you in this talk page: once a client has fulfilled the inclusion criteria via third-party secondary RS, you can use primary sources to cite its difficult-to-source features. --Enric Naval (talk) 09:48, 29 April 2011 (UTC)

I find it interesting that you bring up ircII...because I'm the one who last worked on that part of the article. [5]

While your statement of "for example, it's not citing the source itself, it's really citing the NOTES and UPDATES files" is incorrect, you do have it half correct. For the statement "The CTCP protocol was implemented by Michael Sandrof in 1990 for version 2.1.", the 'NOTES' file supports "CTCP" being added in version 2.1 and 'source/ctcp.c' gives us the author, "Michael Sandrof". Similarly, for the statement "The DCC protocol was implemented by Troy Rollo in 1991 for version 2.1.2.", the 'UPDATES' file supports "DCC" being added in version 2.1.2: "Direct Client Connections implemented. /DCC CHAT opens a direct connection to another client. [...]" and yet again the 'source/dcc.c' file gives us the author, "Troy Rollo". ircii-2.1.4e.tar.gz While neither of these facts is disputed, and can be backed up with other sources, this is an example of where a primary or self-published source is both reliable and appropriate.

Now, lets take "CTCP ACTION" (aka the /me command) in ircII-2.1.5 as another example. While the 'source/HISTORY' file states "Added commands /me and /describe to produce ACTION messages." and "Added CTCP ACTION handling.", as a hypothetical example we could cite 'source/ctcp.c' by itself to support a statement such as "ircII supports the "CTCP ACTION" event". ('source/ctcp.c' also gives us the name of the author, lynX, who anyone familiar with early IRC history will probably know as Carl 'lynX' v. Loesch.)

As for citation templates, {{Cite IETF}} was a red-link too, until I created it (yes, the entire thing, subtemplate "functions", docs, regression tests and all). Previously we used (and still use) a combination of {{Cite web}} and {{Cite book}} for IETF documents. --Tothwolf (talk) 03:38, 30 April 2011 (UTC)

About names of authors, I already told you that it should be OK to cite significant comments inside a source code file.

About citing from the HISTORY documentation file and then citing also the source file, that's for purposes of verification, I think that's called "substantiation of what secondary sources say by citing primary sources". For example, in historic articles you can cite what a historian says about a famous general and then you add some primary material like contemporary texts of Plutarch to substantiate it. In a folklore article you can cite from secondary sources that a certain lore character was drawn in a certain way before 1930, and provide drawings pre-1930 to substantiate it (see Talk:Santa_Claus#Coca_Cola). (The fun part comes when secondary and primary sources are in disagreement and you have to asses their relative reliability. This is usually solved by finding better secondary sources that explain why the primary sources are wrong, misleading, or why they need to be interpreted in a certain context. This is very common in ancient and medieval history articles, for example Talk:Alexander_the_Great#A_few_additional_comments.)

Going back to names of authors. If there are other sources that are more secondary, then you should cite those sources. And then you can cite too the primary sources to substantiate the claim, and to provide further verification that the claim is indeed valid.

TL;DR: you can't make a citation that only says "I read the dcc.c file myself and I can tell that it performs DCC ACTION because I can interpret source code". And you should use more secondary sources whenever possible. --Enric Naval (talk) 08:55, 30 April 2011 (UTC)

Once again, I fail to understand how somebody who only has a passing knowledge of the subject matter is able to exercise a line item veto based on what they consider notable or not notable. There are other clients that you seem to have an issue with to, including IRCjr which used to appear in this comparison and was deleted.

Just because you have not heard of a client doesn't mean you have the right to prevent others from seeing it in a comparison table. And I'm willing to bet that you don't have a DOS machine available, so you have no inclination or desire to ever hear about clients for that operating system. (And that applies to any other operating system that you personally have not used.)

Why on earth are we trying to justify inclusion of particular subject matter to somebody who isn't interested in the subject matter, and isn't being very objective? Nobody is asking for separate articles, and that is where the notability "GUIDELINES" would be applicable. MBBrutman 08:11, 22 April 2011 (CST)

I am being entirely objective in asking for a standard inclusion criteria that applies to all clients. If I wasn't interested in the subject area at all I wouldn't be trying to improve the article. IRWolfie- (talk) 21:42, 25 April 2011 (UTC)

"I am being entirely objective^{[citation needed]} [...]"

I'm not sure how you can claim that you are being objective when you admittedly don't know the material: "I admit that I don't know much of anything about these clients, but it is for those who wish to have them included show why so, not for me to justify their exclusion." [6]
...don't care to know the material: "Also I have no interest in answering this colour question you refer to: since I would be about 6 to 8 at the time." [7]
...and continue to try to force through something using the notability guideline as a means to limit article content "multiple 3rd-party RS for entries with no articles" [8]
...even though it is counter to the notability guideline's usage instructions and other core policies. --Tothwolf (talk) 04:08, 26 April 2011 (UTC)

WTF? Everybody, goddamnit, stop attacking editors. Read WP:TIGERS. WP:CIVIL doesn't say pretend to be civil, it says, "be civil." Inflammatory rhetoric does nothing to improve the situation, such as: "continue to try to force through something using the notability guideline". Well, I see an effort to force in content that's so non-notable, so poorly documented, so little-discussed, and so unreliably sourced that the source code is being resorted to, just to claim features which, gee, too bad, can't be verified without original research (compile, verify). Get real: any Wikipedia editor can be objective about whether sources exist, and whether material is properly sourced, regardless of the topic area and expertise. If an editor cannot verify a claim in an article against cited sources, or finds the sources inexcusably weak, they have the incontrovertible right, under explicit policy, guideline, essay, and consensus, to speak up, and have their concerns honored, civilly. It is editors, by consensus, who determine article content - there are many, many editors here who want better sourcing, and inline citations, for claims made in this list cum table. Best to get along with them, and try to build consensus. Yes, compromise is probably necessary. As for why inclusion should be justified, and why defend challenged material, read WP:BURDEN. As for why sources _do_ determine WP content, and should, I leave you with this

If Wikipedia had been available around the fourth century B.C., it would have reported the view that the Earth is flat as a fact and without qualification. And it would have reported the views of Eratosthenes (who correctly determined the earth's circumference in 240BC) either as controversial, or a fringe view. Similarly if available in Galileo's time, it would have reported the view that the sun goes round the earth as a fact, and Galileo's view would have been rejected as 'original research'. Of course, if there is a popularly held or notable view that the earth is flat, Wikipedia reports this view. But it does not report it as true. It reports only on what its adherents believe, the history of the view, and its notable or prominent adherents. Wikipedia is inherently a non-innovative reference work: it stifles creativity and free-thought. Which is A Good Thing. --WP:FLAT

--Lexein (talk) 02:45, 29 April 2011 (UTC)

As I said I do not know the details of every client, nor should I be expected to. IRWolfie- 13:36, 11 May 2011 (UTC)

Continuing

There are still these clients without third party RS: nexIRC, Neebly, KoolChat, DMDirc and ii. --Enric Naval (talk) 19:26, 6 May 2011 (UTC)

And there will likely be quite a few more when the mass-blanking of this article is undone soon (including quite a few which are now blue links and should have never been removed from these tables in the first place)...that is, until people begin restoring and adding references per WP:V. --Tothwolf (talk) 03:43, 7 May 2011 (UTC)

Not to mention, many clients, including nexIRC (like leetIRC noted below) were merged here per AfD. --Tothwolf (talk) 03:47, 7 May 2011 (UTC)

Just to be clear. I don't oppose that more clients are added, as long as they have articles, or there are sources that fulfill the inclusion criteria. --Enric Naval (talk) 09:18, 7 May 2011 (UTC)

We won't be adding more new clients to this article yet, just restoring those that were mass-blanked, which also included the removal of refs and some footnotes (although I am considering going ahead and adding WSIRC as discussed above). Lexein and Yarcanox had offered to help add sources, but we still need to not bulk remove material if a source can't be located right away since Google Books doesn't index everything and since a lot of material is difficult to find today, such as has been discussed elsewhere on this talk page.

As I mentioned to Lexein the other day, ultimately the best thing we can do with a handful of clients which are difficult to source will be to just comment them out for the time being. This will prevent others from wasting a lot of time re-adding them again (often with less accurate information) and will allow us to restore them as sources are located.

SchmuckyTheCat's trolling/griefing [9] on the other hand is not helping to improve this article and if need be I'll post screenshots of me having to ban him from some other sites and communities prior to him suddenly popping up here. I guess he has nothing more to do since ED no longer exists. I've undone [10] his removal of ii per this discussion below. --Tothwolf (talk) 11:53, 7 May 2011 (UTC)

Commenting them out would be ok, while you find sources. --Enric Naval (talk) 17:42, 7 May 2011 (UTC)

continuing again

Schmucky removed DMDirc, and I have commented out Koolchat, Neebly and nexIRC (forgot to mention that one in the edit summary). ii seems to have passable sources now.

After those changes, I am not spotting any further problematic entries in the list? --Enric Naval (talk) 10:37, 29 June 2011 (UTC)

Limechat sourcing is not sufficient. HydraIRC and nettalk should have sources added to their articles by those familiar with them, or be removed. Several clients do not have entries in every table, but that is maintenance. SchmuckyTheCat (talk)

I made a separate section for Limechat. --Enric Naval (talk) 17:45, 29 June 2011 (UTC)

Thanks. --Lexein (talk) 18:25, 29 June 2011 (UTC)

Limechat sourcing, for this list/chart, is sufficient with its significant mention in a reliable source (a book) + one informative primary source. Simple verifiability and reliability are enough. Tothwolf disagrees with insistence on external verifiability, and SchmuckyTheCat disagrees that one independent RS is enough for a list entry : in my opinion, these two extrema cancel out.

Deletion of content is not warranted - commenting out is much preferred, SchmuckyTheCat. And by the way, per the full text of WP:BURDEN, you would have noticed that DMDirc has a primary source, and a descriptive entry at IRCJunkie.org, which has been established (without dispute) as a reliable source on the subject of IRC. Your deletion has made the work of ref'ing that entry that much harder. --Lexein (talk) 18:25, 29 June 2011 (UTC)

Limechat sources

The only two sources for Limechat in google books are:

Beggining Ruby: From Novice to Professional. p 108 "Mac OSX: LimeChat (http://limechat.net/mac/)—LimeChat was written using Ruby— it's a great demonstration of a Ruby-powered GUI app." looks like a list of Ruby software.

Taking your Iphone 4 to the Max. p 91 "Several IRC apps are available in the App Store that work on your Iphone. Colloquy (...) as is LimeChat (http://limechat.net/iphone: $4.99). Two other apps with loyal following are (...) IRC may be an ancient (in Internet years) technology, but with apps such as Colloquy and LimeChat for the iPhone, it will live for many years to come."

I'm not sure that this is enough to show that Limechat is a notable IRC client. --Enric Naval (talk) 17:08, 29 June 2011 (UTC)

Don't say "notable" - notability per se is not used as a criterion for inclusion here, but reliability and verifiability are. I think Limechat has exceeded the threshold of inclusion for a list/chart: one reliable independent source, and one informative, reliable primary source. --Lexein (talk) 17:49, 29 June 2011 (UTC)

WP:NNC says "(with the exception that some lists restrict inclusion to notable items or people)". We just had a RfC to set an inclusion criteria in this list, We sort of agreed that each entry needed a mention in one independent RS. I think that these two mentions are not substantial enough to qualify. Can anyone find better source(s) for this client? --Enric Naval (talk) 18:21, 29 June 2011 (UTC)

Nah, we aren't going to have this out again, Enric. You already know better. You previously tried this argument here, and here, among other places. It has been repeatedly rejected by other editors even as far back as this. Your prior argument also included WP:LSC, that is until it was discovered that the Manual of Style page which WP:LSC links to had accidentally been tagged as a content guideline (along with a number of other such pages) when it was meant to be tagged as a MoS guideline... You dropped that argument when that was pointed out and corrected but then went on to try to make other arguments to exclude content.

This also isn't a list, this is an article which includes tables of facts, as discussed at length elsewhere on this talk page. This would be a list of it was simply a list of wikilinks to other articles, which is what WP:LSC was written for. As also previously discussed on this talk page, we simply didn't have comparison articles when that Manual of Style page was being drafted. Now, the text you mentioned at WP:NNC as currently written could apply to a list such as User:Tothwolf/List of Internet Relay Chat clients, if that were previously agreed to by the editors who maintained it. When comparative elements, footnotes, and text are introduced however, it ceases to simply be a "list of wikilinks" and "The notability guidelines do not apply to article or list content" part of WP:NNC becomes the rule. While it would make no sense to create a list of nothing but redlinks, which is what WP:LSC was aimed at preventing, neither it nor WP:NNC were ever intended to apply to other article content, just long lists of wikilinks.

Something else to take note of is the very top of WP:LSC itself still links to Wikipedia:Notability (people)#Lists of people which in turn links to Wikipedia:Stand-alone lists#Lists of people. The reason for this is simple, the longstanding wording at WP:NNC used to explicitly state "[...] except for lists of people" and linked to the notability guideline for people. This more recent "addition" to WP:NNC is still disputed. Does the name Gavin.collins mean anything to you?

Enric, this is a clear case of WP:STICK, please drop it. --Tothwolf (talk) 19:40, 29 June 2011 (UTC)

Please show where WP:NNC is disputed, or dispute it yourself in the policy's page. As far as I am concerned, most editors agree that list articles can have their own inclusion criteria to limit the number of items, and that the inclusion criteria can require a certain level of notability for items in the list. For example here, here or here. This is also supported at WP:LISTS#List_naming: "Instead, the lead makes clear whether the list is complete, or is limited to famous or notable members (i.e., those that merit articles)."

The Manual of style says quite clearly that a table with sortable columns is a list article, see WP:STANDALONE#General_formatting and WP:LISTNAME. The table format is mentioned in MOS:LIST#Tables, and the see also section links to a page on how to make sortable columns.

While looking for Gavin's ban rationale, I found the "inclusion criteria for lists" RfC. It's about the notability of the lists themselves, but there is some applicable stuff like ""List of..." is irrelevant. A stand-alone list by any other name is still a stand-alone list.", which is coherent with WP:LISTNAME.

(The name Gavin Collins is vaguely familiar to me, we have commented in the same discussions a few times. I had no idea that he had been banned. No, I don't think that I am going to get banned for my comments in this talk page. Now, you, on the other hand.....)

Now, please, your time would be better spent finding sources for Limechat and other IRC clients. --Enric Naval (talk) 21:02, 29 June 2011 (UTC)

I'm quite satisfied that the existence, and indeed recommendation, of Limechat has been independently verified by at least one independent reliable source, and its primary source is informative enough. That's enough. It's no longer unsourced. Battling for more is the same as battling that an article must exist for every entry, and we already agreed that that is not the way forward for this article to be inclusive without being indiscriminate. We, as editors, should simply recognize when sources have spoken. The problem is solved. Our decision is made for us by the existence of those sources. I believe in a balance between the dual Wikipedia goals of informativeness and reliability and verifiability, and I think that is accomplished by compromising, in this way, here, for this article. 1RS1PS. --Lexein (talk) 06:10, 30 June 2011 (UTC)

That's why I asked that the inclusion criteria was one good source, a source that actually talked abut the client and described it. Now we have this entry, which is supported only by mentions in lists of software. --Enric Naval (talk) 10:22, 30 June 2011 (UTC)

"Mac OSX: LimeChat (http://limechat.net/mac/)—LimeChat was written using Ruby— it's a great demonstration of a Ruby-powered GUI app." is not enough. Above Lexein called this a significant mention defending it. Not. Does this also contradict the table itself which says Limechat was developed with Objective C? What is the source for Objective C? SchmuckyTheCat (talk)

It's RubyCocoa - fixed, and confirmed with an additional ref. Schmucky, I can see your objection to the book reference length, but in my beard-stroking opinion, it reads as a recommendation, by an expert, in a book, even without a full review. Since the primary source is fully informative, I'm inclined to pass this entry. Obviously, more exhaustive reviews are desirable, but that desire, to me, does not exclude this client. --Lexein (talk) 21:19, 6 July 2011 (UTC)

No more deletions. Comment out, please

This is a formal request that there be no further deletions from tables - instead, enclose disputed items in hidden comments. Deletion imposes too great a burden on re-adding editors once the revert window has closed. There has been consensus that all entries without articles will be sourced, so deletion is not necessary - commenting out is enough. (Enric, thanks very much for commenting out, and assuming good faith on future sourcing.)

Can we get consensus on don't delete, just comment out, please? --Lexein (talk) 00:21, 7 July 2011 (UTC)

Neebly

I think that a review in a download website is a bit too far-fetched for "a reliable third-party source". --Enric Naval (talk) 00:00, 13 July 2011 (UTC)

Well, they do a lot of reviews there, so there's some expertise. The prose is usually balanced, not glossing over deficiencies, listing pros and cons. Though they are on the short side (this one's 269 words), reviews there seem to have a consistent style and format. (I don't like that a reader has to click "read more" to see the whole review.) Anyways, since that's the only independent source at the moment, if editors feel strongly about it, I wouldn't mind re-commenting it. Thoughts from others? --Lexein (talk) 01:50, 13 July 2011 (UTC)

I've posed the question at WP:RSN#Softonic.com - if thumbs down, I'll re-comment the entry. --Lexein (talk) 06:40, 13 July 2011 (UTC)

RSN archived here. Well, meh, it appears that people think that Softonic reviews are reliable.... --Enric Naval (talk) 07:07, 28 July 2011 (UTC)

IMHO, only if used with care. Editor discretion is still required. In this case, the Neebly review was written by their most prolific reviewer, who seems to have written reasonably high-quality, informed reviews so far. If it had been a nonsense review, I would never have tried citing it. --Lexein (talk) 12:06, 28 July 2011 (UTC)

Emacs based clients?

Why aren't ANY of the emacs-based clients included? There are at least 3 notable ones: Zenirc, ERC, and Rcirc --24.6.228.145 (talk) 20:32, 27 July 2011 (UTC)

Zenirc doesn't have a page, a search produces a lack of independent reliable sources for inclusion here. ERC (IRC client) does have a (now) well-sourced article, so it could reasonably be included. Rcirc does have an article, but it's very stubby at the moment. --Lexein (talk) 12:11, 28 July 2011 (UTC)

So ERC is in, rcirc is ready to add, but I'm not aware of its features. Oh, and Circe is for emacs too! --Lexein (talk) 13:12, 2 August 2011 (UTC)

The Zenirc website itself isn't a reliable source for its attributes? I count eight clients in this comparison which don't "have a page" at all, and I didn't even bother to see how many of the rest of them had "stubby" pages. That just isn't a good argument against inlcuding clients in the comparison. "Notability" for the clients I named should also be easy to establish, at least to the standard represented by certain other clients in the list. Anyway, maybe I'll get around to boldness instead of mouthyness. --24.6.228.145 (talk) 20:13, 4 August 2011 (UTC)

I didn't say Zenirc's site isn't reliable for features. I said there are no independent RS at the moment.
We have, for the eight clients without articles, included them here with inline citations of a primary source and at least one independent substantive WP:RS, in the topmost table, lefthand column.
My remark about "stubby" articles was simply that we prefer an article to be substantial. The only reason I haven't added rcirc yet is that I get hives editing tables. PITA. I added ERC, and got hives.
BTW the RS doesn't have to support all features claims, but it does have to be a non-trivial discussion of the client, not just a listing, or a terse comment. This is why Circe and Zenirc aren't quite ready for listing here yet. I say "yet", sir.
Correction: Notability is not about the content of an article, it's about the notability of the topic of an article, so we don't say "notability" here about clients, we use "verifiability" and "reliability" of sources. I'm shocked, shocked I say, that you haven't pored through every word of the above discussions (heh). --Lexein (talk) 21:11, 4 August 2011 (UTC)

rcirc has been added, and ERC expanded. --Lexein (talk) 16:06, 5 August 2011 (UTC)

Use of rh vs rh2 templates?

Both {{rh}} and {{rh2}} are in use. Any rationale? I'm tempted to standardize on {{rh}}, if nobody objects. --Lexein (talk) 21:41, 12 July 2011 (UTC)

{{rh2}} has historically been used in software comparisons to indicate software which is no longer being developed and/or maintained. --Tothwolf (talk) 10:07, 18 September 2011 (UTC)

Ah. That's helpful, and belongs in a hidden comment at the top of articles where used. One question: what's the threshold for "no longer maintained?" A primary source on the project? I'd be cool with that.--Lexein (talk) 10:57, 18 September 2011 (UTC)

Mainly common sense. With a client such as PIRCH or Homer, there is no doubt that the client is no longer maintained. --Tothwolf (talk) 11:13, 18 September 2011 (UTC)

There used to be text in the article which explained the convention for readers, but it looks like someone removed it. --Tothwolf (talk) 11:18, 18 September 2011 (UTC)

IRCjr

mbbrutman has added back IRCjr under the rubric of MTCP, using the same independent source as LeetIRC. I still wish IRCjr had some standalone mention by itself as an IRC client. This is a comparison if IRC clients, not TCP stacks. Why is that so freakin' difficult to understand? --Lexein (talk) 06:48, 13 September 2011 (UTC)

User:Cameron Scott disputed the addition of IRCjr, and deleted, with the edit comment "Seems NN". This establishes one editor's firm objection. Per the wishes of several adding editors here, and the assent of two (?) deleting editors, I restored and commented out IRCjr pending discussion by involved editors.

It should be observed that while WP:Notability per se doesn't apply to content, only to whole articles, reliability and verifiability (and in this case, applicability) of sources do apply.
My own objection above was objection for discussion, not accompanied by removal. I do want IRCjr to be here, as the subject of a source. It is not: IMHO, FreeDOS helps, but it's not a review, or sole mention of IRCjr. Does the group feel the same way? --Lexein (talk) 09:42, 13 September 2011 (UTC)

There is nothing "freakin' difficult to understand." IRCjr has its own end user documentation. It has its own web page and has had it continuously. It was the first (and only) application I released, back when mTCP didn't have a name and it was entirely embedded in IRCjr.

IRCjr has exactly the same sourcing as LeetIRC - weak references on FreeDOS and the author's home page. The only difference is that IRCjr is part of a larger group of applications and gets downloaded with that group. I can't prohibit people from talking about the group of applications (which includes IRCjr), nor do I want to. Is that really a criteria for excluding it? Am I supposed to change how it is distributed to get it included back here?

Jim Hall's content says nothing more meaningful about LeetIRC than it does about mTCP, which he points out in the announcement includes IRCjr. You (Lexein) specifically mentioned that you considered Jim Hall to be a reliable source.

If it has the same sourcing as LeetIRC, then they should either both stay or both go. Mbbrutman (talk) 12:45, 13 September 2011 (UTC)

If you _really_ think that a mention of a collection is as direct as a mention of a specific client, okay, that's your opinion. I distinctly recall advising you to get the client reviewed or mentioned. If other editors agree about the collection, then it's in. So far, though, I don't know if you noticed, but there's an actual deletion dispute with revert warring going on, as opposed to hiding in the interim. There are bigger problems, is my point. --Lexein (talk) 12:56, 13 September 2011 (UTC)

Lexein, by your reasoning if I write an operating system and an office suite, and then I include the worlds best IRC client in there, that IRC client would be excluded. I'm playing by your rules here - the reliable sources that I cited are the same as those used by LeetIRC, and those don't count as proper reviews of the LeetIRC client either other than to mention that it is shipping with FreeDOS or was updated.

As for the collection argument, I can't force people to talk about IRCjr in isolation, nor is that reasonable. I'm not going to change the way I distribute my IRC client for inclusion in this table - that is not reasonable either.

It effectively has the same sourcing as LeetIRC. They should both either stay or both go.

You can fight with the deletionist. I think that the overly restrictive notability criteria being invoked here are equally damaging. Mbbrutman (talk) 13:24, 13 September 2011 (UTC)

'The Deletionist' - does that makes you 'The SPI'? --Cameron Scott (talk) 13:27, 13 September 2011 (UTC)

Lexein, I think we need to have an external review here. My reliable sources are nearly identical to those of LeetIRC, and the FreeDOS.Org link is not a review, just a listing. To be consistent you should add mTCP, or decide that LeetIRC and others in the tables are not well enough sourced and strike them until they are well enough sourced.

Personally, I think the requirements for this page are too stringent. I am not alone here - it should be reviewed at a higher level by an impartial party. Mbbrutman (talk) 17:56, 13 September 2011 (UTC)

I think I made it clear that I would abide by group consensus on IRCjr. Getting inclusion criteria here has been a hard battle, balancing Wikipedia's requirements for reliable sources and topic-area source availability, quality, and reliability. WP:RS, read narrowly, without regard to expertise and reliability in this topic area, would result in the deletion of most IRC articles, and the drastic shortening of this page. The criteria here, under the extant fragile consensus, should not be dismissed so easily as "too stringent." Things could be so much worse. I do wish you would read, and assume good faith about (and appreciate), all the hard work that has gone before to save this and other IRC-related articles from oblivion, while still supporting the Wikipedia tenet of independent reliable sourcing of all content. --Lexein (talk) 02:12, 14 September 2011 (UTC)

Please stop with the patronizing lecture and stick to facts.

Let's narrow the scope of the discussion - reliable sources. IRCjr has the virtually the same sourcing as LeetIRC. The only difference is that it is distributed with a group of other programs. The reliable source used in both cases are the home page for the programs and a reference from FreeDOS. The reference from FreeDOS in both cases note that the programs are available in FreeDOS. That's all. It is not a review.

Given that, how do you exclude one but not the other? Excluding it because it is packaged with other programs is not reasonable. Especially since the supposed 2nd reliable source from FreeDOS.org does not review either program!

Apply the criteria consistently.

As for concensus, Tothwolf was clearly of the opinion that it should be included. I am too. Other people are commenting (anonymously, but still commenting) about the overly restrictive inclusion policies in this article. We don't have consensus - we have a draw, and it is easier to undo than it is to add so the overly restrictive interpretation of the policy keeps prevailing. Mbbrutman (talk) 02:32, 14 September 2011 (UTC)

Mbbrutman (talk) 02:32, 14 September 2011 (UTC)

As I said, I would abide by group consensus specifically about MTCP and the FreeDOS cite. I never accepted MTCP as an IRC client, and I doubt Tothwolf did. The client has never been reviewed or significantly mentioned on its own. When the interested editors return, we will of course continue to discuss it. You decided to misread my suggestion to get IRCjr reviewed as "get MTCP reviewed." That was your choice, and you should accept the (perhaps momentary) consequences of that decision. Please stop using provocative, uncivil language, like "patronizing" when I am only trying to remind you of the work which you wish to dismiss. Please read WP:TIGERS. I think you know that if the inclusion criteria were relaxed, then the article would be decimated again by deleters asserting "unsourced", and afflicted by edit wars again. Period. I do not want that. At the moment of this writing, IRCjr is commented out, not deleted, pending group discussion of MTCP vs FreeDOS. I do not have a personal axe to grind, and I understand that you see an apparent paradox in my position about displaying LeetIRC (a client), and not displaying MTCP (a suite). Sorry. I can only hope that WP:3O takes some time to understand what the fragile consensus arrived at so far has achieved. If you wish 3O to be able to do their job, you and I should stop talking now, and just wait.

Please note that we do not exactly have a draw, User:Cameron Scott boldly deleted IRCjr, but discussed, and now appears to be willing to wait for 3O, for which I am appreciative. Because 3 editors were involved you may have to request assistance at WP:EA, since 3O might demur. I am willing to abide by EA, as well, in case that is not already clear. --Lexein (talk) 03:02, 14 September 2011 (UTC)

That is the problem inherent here - this isn't about mTCP, it is about IRCjr. IRCjr is the IRC client that was listed in this article, deleted, and can not get added back in. The fact that it is part of a collection of programs called mTCP is irrelevant!

I'm not asking anybody to include mTCP - I am asking to have IRCjr included. Your interpretation that IRCjr does not exist because it is part of a set of other programs (all of which are stand alone executables) is ill-informed and misguided.

And you conveniently choose to ignore that the reliable source for FreeDOS that is used to justify LeetIRC being on the list is content free, and is not a review of LeetIRC. The references from FreeDOS just say that it was shipped with FreeDOS. If FreeDOS ships the entire mTCP suite (which is does, as the reference says) then it is shipping IRCjr as well as LeetIRC. Yet that does not seem to be good enough. Apply the criteria consistently.

We are not reaching a consensus - see the part about "stalemate" above. Tothwolf clearly did accept it. It was a part of the article for a good long time before the damaging deletion frenzy.

I filed a WP:3O for this anomaly, but because we have more than two editors involved that is not going to resolve it. Mbbrutman (talk) 03:23, 14 September 2011 (UTC)

Response to third opinion request (regarding the inclusion of IRCjr in this list):

I am responding to a third opinion request for this page. I have made no previous edits on Comparison of Internet Relay Chat clients and have no known association with the editors involved in this discussion. The third opinion process is informal and I have no special powers or authority apart from being a fresh pair of eyes.

I've read the above discussion, browsed recent editing history, and looked at the relevant external sources. My opinion is that IRCjr should not be included in this list. When adding an entry to a list on Wikipedia, the entry must satisfy the inclusion criteria for the list. Since this list doesn't specifically define its own inclusion criteria (a separate issue which should be resolved), it defaults to inclusion criteria of passing WP:V. Obviously, any IRC client which already has its own wiki article should already pass WP:V easily, so they get a free pass. Other clients, however, need to have at least a minimum of third party coverage to be verifiable. There are two sources in the article (currently commented out) for IRCjr. The first is a link to the FreeDOS website, to a page on MTCP. There are zero mentions of IRCjr on this page, so this page verifies nothing. The second source is to the developer's website. This is a primary source, and WP:OR clearly tells us, "Do not base articles and material entirely on primary sources." Since the first source isn't actually a source for IRCjr, the second source is the only source. Unless an indepenent (i.e. third-party) reliable source can be provided on IRCjr, it cannot be shown to pass WP:V, and therefore should not be included in this list.

Mbbrutman also repeatedly contends that LeetIRC has the same sourcing as IRCjr, and it is allowed to remain on the list. Firstly, since LeetIRC's entry on FreeDOS is actually about LeetIRC (and mentions it by name), the sourcing of the two articles is not identical. Secondly, using the WP:OTHERCRAPEXISTS argument on Wikipedia usually doesn't get you far (see also WP:Pokémon test).

Finally, I'd like to remind Mbbrutman that Wikipedia is not a means of promotion, and shouldn't be used in that fashion. The fact that you are (presumably) the developer of IRCjr means you have a conflict of interest, and the fact that the only edits you've made on Wikipedia were to add information about IRCjr means that you are a single-purpose account. Both of these facts seriously harm your credibility here when it comes to discussing things related to IRCjr. This may seem counter-intuitive (after all, who is a better source of information about IRCjr than the person who created it, right?), but this is the way Wikipedia works. Wikipedia is a tertiary source, we don't get our information directly from the creator of something because we are not publishers of original research. If IRCjr was notable, then an independent reliable source would have already written something about it and we wouldn't be having this discussion. Furthermore, conspiring to "astroturf" Wikipedia is not constructive, and I assure you it won't work. —SW— ^squeal 23:15, 16 September 2011 (UTC)

I had not noticed that the FreeDOS MCTP page didn't even name IRCjr explicitly. About the astroturfing: it was my suggestion that Mbbrutman "get IRCjr reviewed". He referred to the practice as "shilling" and "laughable". I reasonably expected him to conduct that process via email, not publicly, with "I want to astroturf Wiki", as an administrator of vintage-computer.com. Wow. --Lexein (talk) 00:42, 17 September 2011 (UTC)

Advertising? You are kidding me, right? This is open source DOS software ... which is why we are having such a problem with notability. Nobody is reviewing DOS software; the notability guidelines are inappropriate for such a narrow domain. This point has been gone over and over ... it narrows the scope of the article so greatly and it's not a credible article. Look at the count of clients - 30 or so. Some of them have not been updated since the last century! Can you really call this a "Comparison of IRC Clients" given how incomplete it is?

Getting to the specifics of IRCjr, one of the references that you both failed to mention - a FreeDOS.Org announcement mentioning IRCjr as part of the mTCP suite. (http://sourceforge.net/news/?group_id=5109&id=302479). That reference was also in the deleted (now commented out) IRCjr entry; look at the version history footnote. It was a recent reference too. So yes, the FreeDOS.Org owner is aware of it and mentions it by name.

As for conflict of interest, I am the author and I've never hidden that. Authors of other IRC clients are trying to get their clients added to the list too, *for completeness*. Right now you have one IRC client for DOS listed; there are at least three. I can assure you when the author of that one lonely client added it to the table nobody accused him of a conflict of interest. I am quite frustrated with the intransigence here to opening the article up, and excuse me if it shows. Nobody really expects to be able to force anything in, because the (undo) button being wielded here by the people sitting on the article is too quick and too easy. You do not see any bogus editing caused by that forum thread - learn to take things in context. Mbbrutman (talk) 03:45, 17 September 2011 (UTC)

SW - one more thing before this matter is closed. You accuse me of being a single purpose editor. Go see the IBM PCjr page - I clearly am not a single purpose editor. There are also plenty of other edits that I did anonymously to deny people the chance to carp; not every editor needs to be on an ego trip and see their name in print. Mbbrutman (talk) 04:22, 17 September 2011 (UTC)

The notability guideline refers to whether or not a certain topic can have it's own full article. It should not apply to something like this. This page is supposed to compare IRC clients. If you want to exclude certain clients that clearly exist because they are not popular, this article's title should be changed to "Comparison of popular IRC clients"... otherwise, leave them ALL if they are verifyable. And this is IMO, but I think IRCjr and LeetIRC are particularly notable because of the fact that they are clients for DOS that are still under development. That is a rarity, and they certainly will have their niche of users even in 2011. They appear to be the only two. 24.107.102.129 (talk) 02:53, 19 September 2011 (UTC)

With all due respect, an IRC client isn't notable just because you (or I) "think" that it is. If they are so unique because they are DOS clients that are still under development, then someone other than their authors would be writing more than 1-sentence mentions about them. List articles always have inclusion criteria, just because this one doesn't specifically define the criteria doesn't mean we can or should include every last IRC client that has ever existed. A line has to be drawn somewhere. —SW— ^express 05:03, 19 September 2011 (UTC)

A gentle reminder, and welcome to a possibly new user. First: Wikipedia WP:Notability isn't the same as the commonly used term "notable". WP Notability means as reported by multiple reliable independent sources, over a period of time, establishing significant importance. We already know that many IRC clients don't have enough RS to meet the General Notability Guideline, to have their own articles, and may never, and we're at peace with that. That particular alligator has left the swamp. Second: Correct, IP user, WP:Notability applies to articles, not items. We editors occasionally goof and say "notable" when we mean to say "verifiable - discussed in an independent reliable source". I keep saying: we need help digging up sources which aren't online, and reliable sources can pop up in some of the most surprising places: WP:WikiProject IRC/Sources --Lexein (talk) 06:00, 19 September 2011 (UTC)

The notability guideline doesn't have anything to do with the inclusion or exclusion of article content. The only thing which matters as far as article content goes is the verifiability policy. This has already been debated heavily so there is no point in trying to argue notability again.

Now, as far as "[...] doesn't mean we can or should include every last IRC client that has ever existed." goes, because there is a very finite number of clients, we technically could include them all, but there are two reasons why we probably shouldn't... One, due to browser limitations it would be difficult to include all clients in a single article and have it still remain useful to readers (we would have to split the material into multiple articles, which would then limit the article's usefulness as a Comparison of x). Two, there were a lot of small one-off clients written in languages such as Visual Basic which never really had any users and were largely of interest only to their author, and including those would not really be beneficial to readers.

At the same time, the lame edit wars spurred by a handful of individuals who are only here editing Wikipedia in an attempt to seek some sort of self-importance while having zero actual interest in (or even knowledge about) the topic or content they are edit warring over has made it extremely difficult to find balance between including material readers have expressed interest in reading about and including nearly nothing. --Tothwolf (talk) 16:16, 19 September 2011 (UTC)

mbbrutman: stop moaning! I am also an author of a relatively new IRC client (ClicksAndWhistles) which did have a standalone article and entries in the tables but it was deleted due to not being notable enough at the time (it may be now). Editing an encylcolpedia for the purposes of shameless self promotion is just not on! :) --Leigh Preceding comment added by 86.175.41.70 (talk) 22:04, 19 September 2011

It's actually a case of "... not being verifiable in a reliable source at the time ..." --Lexein (talk) 03:20, 20 September 2011 (UTC)

Erm, actually, ClicksAndWhistles was removed from this comparison article as part of that mass-blanking last year in August 2010 and didn't have anything to with either notability or verifiability, see this diff. This took place right after the last AfD attempt by User:JBsupreme/User:Miami33139/User:Theserialcomma (the first two accounts eventually abandoned after being sanctioned by ArbCom and the third later blocked for this very behaviour), which itself was part of their campaign of harassment towards me (since I had been working on this article), and is detailed in the links at the top of my talk page. Basically, the later AfD attempts of this and other articles I had previously edited began to fail (as other Wikipedia editors caught on to what these three individuals were up to), so they changed tactics and instead took to coordination and meatpuppetry on the talk page to give the illusion of "majority consensus" in order to try to blank the article instead. Considering that they were doing this purely "for the lulz", it is rather unfortunate that a few more people got sucked into this and took up their cause, which has resulted in a continuation of the drama and edit warring long after those three individuals were sanctioned. --Tothwolf (talk) 08:20, 20 September 2011 (UTC)

Ok, I just did a check around for sources for ClicksandWhistles, and didn't see anything yet in sources which could be rehabilitated as reliable. At the diff above, ClicksandWhistles didn't list an independent RS discussing it, for verifiability. Back to the present, we're still agreeing about the need for independent verifiability, aren't we? I mean, several clients are back specifically because of our mutual and cooperative work in that regard. --Lexein (talk) 09:16, 20 September 2011 (UTC)

What/who determines if a source is "reliable"? What are the reliability "metrics" exactly? ClicksAndWhistles has been reviewed by Softonic and Addictive Tips as well as many others; are neither of these sources reliable enough and if not why? Do *all* the other clients in these tables have more reliable sources? (Leigh) 86.175.41.70 (talk) 13:04, 20 September 2011 (UTC)

Read WP:RS for a discussion on how to determine if a source is reliable. —SW— ^gab 15:16, 20 September 2011 (UTC)

WP:Identifying reliable sources (WP:RS) addresses the most favored qualities of sources considered reliable. Exceptions and questions are discussed at the Reliable sources Noticeboard (WP:RSN). For IRC articles specifically, two sources have been rationalized as reliable at WP:WikiProject IRC/Sources.

At WP:RSN, specifically here it was noted that at a prior article deletion discussion, Softonic was considered reliable. I'd be inclined to include ClicksandWhistles based on that review. The AddictiveTips review is an open question. I'll ask. --Lexein (talk) 15:36, 20 September 2011 (UTC)

We comment out, not delete, pending sourcing.

Where do you get this idea from? Unsourced content can be deleted at any time by anyone, the onus is on the person adding the content to find source before returning not the person deleting it. --Cameron Scott (talk) 09:43, 13 September 2011 (UTC)

(Question copied from User talk:Lexein)

There is weak consensus here at this article among those involved, both adding and deleting editors. Discussion here encompasses a long history to which you may not have been a party. It's in the archives. Because of the difficulty in adding content to tables, and the request by adding editors, and expressed agreement by ~~at least two involved deleting editors~~one involved deleting editor (and no comment by another), we've arrived at the compromise of commenting out, pending additional sourcing. We've also agreed that primary sources count, and significant mentions in RS count, for those items without articles.
There are sources, they just aren't quite enough for this article's inclusion criteria, IMHO. The content is not literally "unsourced."
WP:BURDEN also says: "It has always been good practice to try to find and cite supporting sources yourself." This is in the spirit of building an encyclopedia. --Lexein (talk) 09:54, 13 September 2011 (UTC)

I'm not big on spirit, policy backs me. We could argue about this but the quickest way to resolve it is simply for you or someone else to find some decent sources for it's inclusion. I never remove properly sourced material from articles. --Cameron Scott (talk) 10:00, 13 September 2011 (UTC)

We agree on more than you think, but I must repeat, welcome to this article. Please read the 2010 and 2011 discussions to see where things are here, now, especially about inclusion criteria, rationale for use of particular sources, and rehabilitation of sources. Policy backs us all. That's why we discuss and reach consensus. Our working consensus here is within policy letter and spirit, intended to improve Wikipedia while not unduly penalizing the editing team, when content which is nearly sufficiently sourced, and can be reasonably expected to be fully so. --Lexein (talk)-

I'm not new to this article - I deleted a lot of content from it last year. --Cameron Scott (talk) 10:24, 13 September 2011 (UTC)

Well, then, welcome back. We've sourced a bunch since then, and improved a bunch of related articles, to boot. --Lexein (talk) 10:35, 13 September 2011 (UTC)

Your deletion and reversion and edit summary goes against policy, against consensus here, and against your inferred consent above to let the commented material stand. You again deleted after further discussion clarifying the consensus of involved editors here prior to your deletion. The material is not unsourced, it has two sources. Its sourcing (arguably) does not yet fully meet this article's inclusion criteria yet. In this case we've commented pending sourcing. The material is not visible to the public, is not controversial, is not BLP, and is not extraordinary claims, so deletion is not called for. I've taken this to WP:3O. Stop until 3O responds. --Lexein (talk) 10:55, 13 September 2011 (UTC)

Third opinion

It is acceptable to comment out a small portion of text while an on-going consensus seeking attempt is going on (see WP:COMMENT and Help:Hidden text). However, here we have a situation where a large piece of text is commented out not while there is a search for consensus but in the hope that a source verifying the commented text will arrive at a later date. This is not acceptable both because of the size of the commented sections and because of the uncertainty that such a source or sources will ever arrive. If there is support for the text, but no reliable sources, then may I suggest storing the commented text in a subpage of the talk page so that it can be easily accessed when necessary. In the meantime, the commented text should be deleted. --rgpk (comment) 13:54, 16 September 2011 (UTC)

What's the definition of "large"? Each item constitutes of a hundred bytes or so of content, excepting templates, table formatting codes, and refs.

OK about deletion, though it's a pain to add back - six tables to be edited, in WP's non-WYSIWYG interface. I wouldn't care at all about deletion vs hiding, given a better editor, and would likely not have ever bothered with commenting. The "stash" subpage helps a little, I suppose, in the interim. --Lexein (talk) 15:32, 16 September 2011 (UTC)

Collectively, the hidden text is very large. Regardless of length, each individual line should not be included anyway because of the uncertainty associated with when a source will be found. Hidden text is generally meant to be used for instructional or informational purposes and is not really meant to park content (except when consensus seeking is in process). I agree that saving disparate table entries in a subpage is a bummer but that's really the proper way to go.--rgpk (comment) 18:47, 16 September 2011 (UTC)

Actually, the size of the text does not matter per Wikipedia:Don't worry about performance. The commented text prevents needless re-introduction of material by well-meaning editors who will otherwise waste a lot of time adding material they "notice is missing" from the tables. The only time the commented material is going to be seen is during editing, so it doesn't get in anyone's way. Commenting out otherwise correct and fact-checked material also allows us to easily restore it later as references are improved. --Tothwolf (talk) 09:59, 18 September 2011 (UTC)

Post 3O

Remove commented items: agreed. Remove item (IRCjr) not named in source: agreed. But since 3O did not declare LeetIRC "poorly sourced", I've partially reverted SchmuckyTheCat's edit. Careful with that trigger finger, Tex. --Lexein (talk) 17:28, 17 September 2011 (UTC)

Once again, the selective editing is mind boggling. IRCjr is clearly mentioned in this other FreeDOS reference (http://sourceforge.net/news/?group_id=5109&id=302479) . Mbbrutman (talk) 17:52, 17 September 2011 (UTC)

This edit, added by you 05:05, 13 September 2011, does not include the Sourceforge reference, and 3O didn't know about it. That's on you. You mentioned it too late, and buried it in a bunch of yelling. Now, let's ask:

SchmuckyTheCat and Cameron Scott: Is IRCjr sourced well enough with FreeDOS "irc", not "IRCjr", primary, and now Sourceforge announcement?

I don't know if such SourceForge announcements are considered reliable.

--Lexein (talk) 18:26, 17 September 2011 (UTC)

For showing existence? Or for showing the minimum level of notability that we should put it in the list? Yes on the first, no on the second. This is also why LeetIRC should be removed. A database listing from an open source project merely shows existence. SchmuckyTheCat (talk)

Do you not read what other people try to edit here? Go to that version. Go to the release history table. Look at the date. Follow the footnote! I had that link on the release history footnote! It's been there all along! That is on you for deleting first without thinking about it.

I specifically tried to make the IRCjr entry as clean and as properly sourced as the LeetIRC reference to avoid this silliness. You yourself said it looked like the same references last week when you first tried to justify deleting it.

It is at least as well sourced as LeetIRC. Mbbrutman (talk) 18:35, 17 September 2011 (UTC)

As for the SourceForge announcement not being reliable enough - don't raise the bar again. The announcement is from Jim Hall. It is project news for the FreeDOS project that he maintains. He controls the news items that get posted there; nobody asked him to do that, or gave him wording to use as I'm sure somebody is going to suggest. It's just as credible as if it had appeared directly as FreeDOS.org. (In fact, it does appear right on FreeDOS.org! (http://www.freedos.org/ - scroll down a little.) Mbbrutman (talk) 18:49, 17 September 2011 (UTC)

Shrug. Doesn't bother me that I didn't see your, rather buried, third source, or even that 3O missed it, because for every other client, all basic verification sources are in the topmost table. Anyways, it's down to the two current deleting editors, now, as I indicated above. --Lexein (talk) 19:32, 17 September 2011 (UTC)

"Shrug." That's all you have to say. I went through the effort to properly do the table, include sources, and to make it consistent with the rest of the table. You deleted it on sight after reading the first line - if you had gone to line two, you'd have seen the third source. Which was appropriate to put as a version footnote, just like the other entries in the table do. It was on the second line of changes that I made! How much work would it have been to look at it?

You spend all of this time deleting and telling me why it doesn't belong in the table, when it clearly is as reliably sourced as other things in the table. And all you have to say is 'Shrug'. Where I come from, when people make a mistake they own up to it.

This entire fiasco shows what is wrong with the editing of this article. Now the fate of the entry (and a related entry which you've been defending) depends on two people who've shown a tendency to delete rather than include. It's not like this table was so large that it was out of hand. Mbbrutman (talk) 19:46, 17 September 2011 (UTC)

Yep, shrug. Cameron Scott deleted "on sight", I reverted and commented it out instead. I have nothing to own up to. We work with what we're given. You did not explicitly mention the Sourceforge ref per se (used only in the 2nd table) until after 3O. Nobody would have missed the Sourceforge ref if you had put it in the first and second table. And you didn't mention the FreeDOS home page with the announcement until just above here. I'm pretty sure I'm free to wait for a consensus of other editors. I may contribute further, I may not. Just wait for the other editors to contribute. Table size was never an issue, by the way. --Lexein (talk) 20:37, 17 September 2011 (UTC)

The burden is on you to read before deleting! That's a basic responsibility of an editor! I had wrongly assumed you did at least that. It's fascinating that any mistake I make is mine, and any mistake that you make is mine too. (Mike) 96.42.66.188 (talk) 21:06, 17 September 2011 (UTC)

No, you! --Lexein (talk) 21:50, 17 September 2011 (UTC)

WP:LAME --Tothwolf (talk) 10:04, 18 September 2011 (UTC)

WP:SHRUG --Lexein (talk) 11:26, 18 September 2011 (UTC)

Absolutely fascinating! You delete edits without fulling reading them. When it's pointed out that the material you deleted is properly sourced (at least by your rules) you scold for not bringing that up early. When it's pointed out that the original edits that you deleted had the reference, you blame it on me for not pointing that out, without taking any responsibility for not reading what you pre-emptively deleted.

And now you are are posting links to funny pictures claiming I'm the problem. (That's an elementary school tactic.) Let's see, an editor who deletes without reading what he is deleting, or somebody who makes an edit with properly sourced material and has it deleted on sight. Where is the problem? I think anybody who is reading this can figure that out, despite your refusal to take any responsibility for a preemptive delete, and then spending a week trying to justify it.

And now you are posting essays to defend yourself. You should spend as much time reading the material you are about to delete! Then maybe this kind of crap wouldn't happen, and Wiki would get such a bad reputation. I love the part about not wanting to hear anything from particular people. Apparently you don't like being wrong, and called out on it. Try to be more professional - you dropped the ball on this. You've been going way over the top to try to prevent this particular edit, even though there are other clients on the list as well (or poorly) sourced. Your bias shows pretty clearly.

It's time to stop the madness. We should be discussing what the inclusion criteria are in the face of a thinly sourced subject area - IRCjr is not the only client that is being blocked from the table. Do we need another table/article for lesser known clients? Is there a better way to have these clients listed without tables that are so wide that they fill the screen, and then some? (Mike) — Preceding unsigned comment added by 96.42.66.188 (talk) 19:11, 18 September 2011 (UTC)

WP:UNCIVIL. As I have repeatedly stated, and the edit history will back me up, I deleted nothing preemptively, and I deleted nothing without apparent consensus. A review of the inclusion criteria discussion will show WP:V, at least one substantially mentioning RS plus informative primary source. FreeDOS was a reluctantly permitted, and on-the-bubble, source held up by (apparently solely by) my personal enthusiasm, and always persistently opposed by at least two other vocal, involved, editors. Said enthusiasm has, as you might have noticed by now, subsided. --Lexein (talk) 19:39, 18 September 2011 (UTC)

You say you deleted nothing pre-emptively. Yet you deleted a new entry, then blamed me for not pointing out a reference that was part of the entry you deleted. So at a minimum you deleted without fully reading, and tried to assign responsibility for the missed reference to me.

Just read before you delete something. It makes things easier. It is also part of the responsibility of editing. (Mike) — Preceding unsigned comment added by 96.42.66.188 (talk) 19:50, 18 September 2011 (UTC)

Mike, the additional source that I didn't mention in the 3O response above really doesn't change things. The source barely mentions IRCjr except for saying that it is the IRC client for mTCP, a TCP/IP stack which is used to run programs on ancient computers. If your entire argument rests on the fact that LeetIRC is sourced as poorly as IRCjr was, then your argument is that other crap exists on Wikipedia. There is nothing preventing anyone from adding non-notable entries to this list, therefore the existence of an IRC client in this list doesn't conclusively prove that it is notable. If you truly believe that LeetIRC is not notable enough to be on this list, then delete it from the list. As long as you do it in a genuine and non-spiteful way, it's as easy as that. But, I suspect that your concerns are about IRCjr, not LeetIRC, so why don't we keep the arguments relevant? Sometimes it's difficult for people to see how something could be considered non-notable on Wikipedia when it seems so notable to them. This is one side effect of having a conflict of interest, it's difficult for you to look at the situation objectively. Consider the types of sources which are ubiquitous for truly notable IRC clients, like ChatZilla or Mibbit, and then look at the sources that you're putting forward, which are data tables and single-sentence mentions. I think your best move at this point would be to drop the stick. —SW— ^comment 21:51, 18 September 2011 (UTC)

I am taking your advice - this is futile.

I was told (in the history of this discussion) to get more sources. I waited for more sources to become available - specifically the FreeDOS crowd picked up on the code when I released the source code. That made the references effectively the same as LeetIRC. People have pointed to LeetIRC as a legitimate entry, and even suggested that I have Mike Chambers (the author of LeetIRC) write a review to give me an additional reliable source. So using the same criteria as applied to LeetIRC seemed perfectly logical to me. The refusal to add IRCjr but leave LeetIRC seems wrong given that LeetIRC was the standard I was given to use.

The discussion here should be what are the inclusion criteria; following the lead of Cnet and Lifehacker is going to cause a lot of clients to be excluded, and needlessly too. This article is not serving its readers with such a narrow focus. (Mike - retired!) — Preceding unsigned comment added by 96.42.66.188 (talk) 23:58, 18 September 2011 (UTC)