Talk:Internet Relay Chat

Image copyright problem with Image:Xaric screen shot.jpg

The image Image:Xaric screen shot.jpg is used in this article under a claim of fair use, but it does not have an adequate explanation for why it meets the requirements for such images when used here. In particular, for each page the image is used on, it must have an explanation linking to that page which explains why it needs to be used on that page. Please check

• That there is a non-free use rationale on the image's description page for the use in this article.
• That this article is linked to from the image description page.

This is an automated notice by FairuseBot. For assistance on the image use policy, see Wikipedia:Media copyright questions. --08:57, 5 November 2008 (UTC)

Unclear how it all works

I know virtually nothing about IRC, and after reading the article, I'm still a bit confused by a lot of the big picture questions.

I infer/assume that a network is composed of servers, and that channels are either local to a server (&) or span across the entire network (#). Channels can not span networks, though (I'm unsure of this). Clients themselves connect to a server, but the choice of server is mostly down to latency and server-only channels, as any server can access any network-wide channel in that network.

Is that right?

205.166.76.15 (talk) 22:45, 6 July 2009 (UTC)

Channels are network-wide, but do not carry over into other networks. So, there is a #help on EFNet and another #help on Dalnet. I'm presuming a bit much, but local channels are rare. The vast majority will be network-wide (#), so I wouldn't get bogged down on the local channels (&) which may only be in place for local IRCops. Clients have a choice over what server they want to connect to. There are some round-robin domains that can connect to you a random EFNet server, for instance, but a client can pick from dozens of servers that are connect to the same network. If the server is too slow (latency), the client can stick it out or choose to switch to another server that may be faster. Rurik (talk) 23:55, 6 July 2009 (UTC)

Some channels may be linked to several networks through the use of janus. For example, the same #net is available on irc.hackthissite.org and 2 other networks. But for the vast majority (~99.99%) of channels, they only exist on the network and are not inter network. Netalarm 10:24, 24 July 2009 (UTC)

This has nothing to do with the actual IRC protocol, however. That is a client-side relay and not notable. TrbleClef ♮ (talk) 20:08, 25 July 2009 (UTC)

A network is a cluster of individual servers. When you want to connect to a network, you choose a server thats best for you based on latency. There are many different networks out there all run by different people. Much like there are different websites. What theyre trying to say I think is that # is a network wide channel, meaning it doesnt matter what server you connect to within the network, all the channels text will be displayed the same on every server. Whereas if it was just local (&), youd only see it if you are connected to the same server, and there could be many of the same channel, with different users on one network. Just on different servers. Its kind of moot to say that it doesnt broadcast to other peoples networks. You can pretty much assume that. It would be like saying that me posting here isnt going to show up on someone elses wikipedia. I guess its just there to confuse people.

Also another thing that I thought was kind of moot is this line

with hundreds of thousands of channels (the vast majority of which stand mostly vacant)

The second part kind of cancels out the importantce of the first statement. Why even include the statement. Might as well just worry about the number of networks and users. —Preceding unsigned comment added by 68.83.71.178 (talk) 16:14, 7 August 2009 (UTC)

RE: Modern IRC unsourced material box: As far as I know, much of this information is common knowledge. Depending on your definition of "modern" (2k+ vs -2k?), sources for the services data can be found on this page. --68.183.131.94 (talk) 19:52, 18 September 2009 (UTC)

Direct Client to Client

"as well as chat and data transfers via Direct Client-to-Client." I don't really think thats a feature of IRC. The on DCC info sent through the IRC server is that someone wants to DCC you (which is done through a NOTICE) JV Smithy (talk) 00:47, 22 December 2009 (UTC).

History

The article seems pretty light on the history of IRC, conspicuously lacking information about the original IRC network (EFnet), the significance of ircII (especially the 2.8.1 release). I understand that finding reliable sources will be a challenge, though.  :( //Blaxthos ( t / c ) 22:16, 3 March 2010 (UTC)

The Standard (RFC1459) modes Section

Can this section be tidied or clarified a little? In the 'User Modes' section, there are four modes which are 'Channel User Modes' (+q,+o,+h,+v), however +o and +v are also listed in the 'Channel Modes' section.

Something else to note there, is that RFC 1459 does not contain the concept of a 'Channel Owner' (+q) or a 'Half-Op' (+h), so for factual accuracy these should be removed, or added to another small section of common non-rfc modes. FrostyCoolSlug (talk) 13:38, 14 July 2010 (UTC)

In addition, the RFC does not state that IRC Operators should have an & prefix. FrostyCoolSlug (talk) 13:46, 14 July 2010 (UTC)

Y Done SpigotMap 14:00, 14 July 2010 (UTC)

Images

Does IRC support images? Can images be transferred over IRC? Is support for display of images dependent on the client? -- Azemocram (talk) 23:52, 29 July 2010 (UTC)

Transferring could be every file. if you can display inline images? Don't think that this is in the specs. Some clients create smilies out of the text-smilies. mabdul 13:57, 13 August 2010 (UTC)

the big four

since you gave a "reference" for the biggest IRC networks shouldn't we change the big four to the actual big four as you partially did? I mean the (our) list is incorrect... (I don't have access to any computer at the moment, will come back constantly in a week or so)mabdul 13:51, 11 April 2011 (UTC)

I assume this was a question for me? It isn't quite that simple. netsplit.de's top 100 is also only ranking by number of users, however if you look at the stats, the number of channels and servers is disproportionately smaller than the other networks. A quick solution would be to multiply the number of users by the number of channels by the number of servers. Taking the results of these numbers, the top 10 works out like this:

Network	Users	Channels	Servers	Average
QuakeNet	79142	59641	42	198244536924
IRCnet	64540	37562	34	82424550320
Undernet	51419	16478	22	18640210204
Ustream	46358	5803	12	5649324954
EFnet	40991	20039	50	41070932450
Rizon	18172	19921	17	6154075004
DALnet	17883	8553	42	6424038558
WebChat	15566	2001	6	186885396
IRC-Hispano	15332	6360	32	3120368640
FCirc	12603	15430	8	1555714320

So you can see how much smaller Ustream.tv's "network" really is in proportion to both EFnet and Undernet. Ustream.tv's proprietary software is also what is being measured, so that alone would allow us to discount that "network". We also have published sources for the "big four", which are backed up with the stats from netsplit.de once the numbers are averaged. --Tothwolf (talk) 15:19, 11 April 2011 (UTC)

Yes keeping the big four as they are makes the most sense, Ustream.tv doesn't seem to be a traditional network and there's no reference to say it has ever been named as one of the big four even if it was larger. IRWolfie- (talk) 21:08, 11 April 2011 (UTC)

I did a little more research on Ustream.tv's IRC support and it's a little interesting. Apparently they run customized servers and have an embedded client called "UStream chat" on their website. A normal client can be used, but it still requires an account on their site and the use of the /PASS command with the account password. --Tothwolf (talk) 21:50, 11 April 2011 (UTC)

Comparism

In recent times of icq, facebook, and msn, I wonder if IRC is still the biggest chat system world wide (in terms of user coverage)? Any figures? — Preceding unsigned comment added by 77.176.228.194 (talk) 17:21, 29 June 2011 (UTC)

Force join

Recently I had an incident of which I randomly joined the #freenode server. I was in chat when the #freenode server just came up. It happened about 6 times before I just left the server open. Can I get an elaboration? Tez011 (talk) 14:31, 17 August 2011 (UTC)

Unclear jargon

The word "opered" is jargon that is not explained, and its meaning is not entirely clear from context. In its first usage, it appears to be a typo for "opened." IAbookman (talk) 14:09, 18 August 2011 (UTC)

Requested move

The following discussion is an archived discussion of the proposal. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.

No consensus to move. Vegaswikian (talk) 02:15, 27 September 2011 (UTC)

Internet Relay Chat → Internet relay chat –

Per WP:CAPS and WP:TITLE: this is a generic, common term, not a propriety or commercial term, so the article title should be downcased. In addition, WP:MOS says that a compound item should not be upper-cased just because it is abbreviated with caps. Matches the formatting of related article titles. Tony (talk) 04:12, 20 September 2011 (UTC)

• Oppose (added post-discussion --Lexein (talk) 15:55, 20 September 2011 (UTC)): No, this is the proper name of a creation, an invention and protocol. The word "Relay" is not generic, it is technically specific to the protocol and implementation, and "Chat" is the specific product of the protocol and implementation. Compare Hypertext Markup Language, which was attacked for similar nonsensical decapitalization, resolved only by renaming to HTML, all caps. There has been ongoing panicky insistence on applying decapitalizarion suggestions (used occasionally elsewhere) as ironclad rules at WP - we should avoid this. --Lexein (talk) 04:51, 20 September 2011 (UTC)

The name of a creation, an invention ... like "telephone"? Just because a thing is an invention doesn't mean it's a proper noun. Is IRC (abbreviated with caps, of course) a proprietary, licenced technology? The lead talks in terms such as "Client softare is now available for every major operating system that supports Internet access". It uses lowercase "n" for "IRC networks", and "s" for "IRC servers". More tellingly, it says, "All client-to-server IRC protocols in use today are descended from the protocol implemented in the irc2.4.0 version of the IRC2 server, and documented in RFC 1459." Aha, so there are many Internet relay chat protocols, not just one. It's a set of technologies. Sounds like your caps are needed for the original one documented in RFC 1459, and that since then the cat's out of the bag and it's like "computer" or "network" ... just a common noun. Tony (talk) 06:37, 20 September 2011 (UTC)

Inappropriate comparison with single-word inventions which have entered the common lexicon. Internet Relay Chat is an atomic thing. There is no Internet Relay Chat Network, though Dave's Internet Relay Chat Network might be a name for a particular network; point is, IRC modifies "network", and modifies "servers" - it is still atomic as a protocol and separate thing. Internet Relay Chat refers primarily to the protocol, and secondarily to the chat, so it is primarily a proper name. We don't say, "I was on the Internet last night, using relay chat", or I was connected to an Internet relay last night, chatting." All three words together form an atomic, unique thing. I would agree with the separate use of lowercase "relay" or "chat" when used as a thing: an IRC server, or a conversation, following the introduction of the full term, w/caps. This follows the example of works discussing Hypertext Markup Language later using lowercase "markup" and "language" as referring to parts of the total concept, but only after introducing it first fully capitalized. --Lexein (talk) 07:43, 20 September 2011 (UTC)

I've never heard about treating single- and multiple-word items differently: this seems to be an ad-hoc, localised innovation to try to sneak this through as capitalised when it shouldn't be.

No one would say, "I was on the Internet last night, using relay chat", because it's unidiomatic. You'd say, "I was using Internet relay chat last night"; even then, we usually abbreviate it, orally, so the examples are slightly awkward. Similarly, an ISP (that's "Internet service provider", not "Internet Service Provider", even though someone originally coined the term) probably wouldn't say to a potential customer that they need an asymmetric digital subscriber line connection; they'd refer to "an ADSL connection". We upcase acronyms and initialisms, but many style guides, including our own, say not to reverse-engineer the upcased abbreviation when expanding the item, unless of course it really is a proper noun. IRC is just a technical term now, given that it's a whole class of technologies, not the original one. Tony (talk) 08:16, 20 September 2011 (UTC)

Calling anything "sneaky" assumes bad faith. Also, you haven't answered my use case, HTML. The problem is not with me, it is with the enforcement of arbitrary capitalization standards which capitalize article names against the wishes of artists and authors of works, and against the proper names of historical items. It is enforcement for the sake of enforcement for no other reason than personal certitude, not in accordance with any external widely applied standards. It goes far too far beyond the lowercasing of "the" and "and." It pains me to see editors, one by one, falling victim to the weird, twisted logic employed to rob all creators of the proper representation of the names of things. It is a laughable situation, and one which damages Wikipedia's reputation, as not even being able to get the names of things right.

To your response: I do not agree that "just a technical term now" applies, given that the name of the protocol and its implementation have not changed, and will not be changed. The fact that colloquially people don't write out "Internet Relay Chat" is not a justification for stripping the proper capitalization from this historical innovation. Before I would rename this article, I would split it - one for the protocol, its history and implementation details, and another for colloquial use.

There has been no "reverse engineering" abbreviation to full name. The name has never changed. Furthermore, that "no reverse engineering abbreviations" is made up and is not supported anywhere outside Wikipedia, it is fabricated for the purpose of pushing an unsupported, and unsupportable agenda. It's ridiculous, and stands in the way of Wikipedia doing its job, starting with getting the names of things right. It should be rewritten from scratch. In the meantime, it should properly be ignored as silly. --Lexein (talk) 09:28, 20 September 2011 (UTC)

As far as I'm concerned, citing "Asymmetric Digital Subscriber Line" -> Asymmetric digital subscriber line above and not mentioning you had just pushed through this requested move is inappropriate. As someone who sees that term on standards documents just about daily, had I seen that requested move, I would have opposed it as well. The WP:TITLEFORMAT section of WP:TITLE explicitly states: "Use lower case, except for proper names" yet this is being ignored in the case of proper names of protocols and standards. It would seem to me we need an explicit exception to this silly MOS rule for protocols and standards because clearly common sense isn't in use. --Tothwolf (talk) 12:32, 22 September 2011 (UTC)

• Oppose. Sources overwhelmingly use uppercase; if MOS rules demand a different capitalization, then MOS rules are wrong. Internet Relay Chat is a proper noun used for a protocol, just like Hypertext Transfer Protocol, Transmission Control Protocol, etcetera. Ucucha (talk) 12:12, 20 September 2011 (UTC)

You know very well that's not the way it works, and that you're sitting there passing or failing FACs is appalling, in that case. Now, is it a single, unique protocol? Tony (talk) 12:24, 20 September 2011 (UTC)

It is a single unique protocol, with subprotocols and commands allowing expansion over time, per the various RFCs. Just like HTTP, TCP. --Lexein (talk) 15:55, 20 September 2011 (UTC)

• Oppose. It is a proper noun. — Dgtsyb (talk) 12:26, 20 September 2011 (UTC)
• Comment. AFAIK IRC refers to an Internet group communications solution that is based on a specific protocol, the IRC protocol. I do not think it refers to Internet chatting in a more general way. I think the burden to prove that there is a general meaning for this term should rely on the person who wishes to lower-case it. My 2 cents, Nageh (talk) 20:55, 20 September 2011 (UTC)
• Support This is not a proper name, just a technical term that is commonly up-cased to match its acronym. The MoS does not support that style. Don't be fooled by acronyms/initialisms - use lower case. Jojalozzo 02:35, 21 September 2011 (UTC)

Oh, please. It's not a "term", except colloquially. It's the title of the specification, protocol, and implementation and has never changed. I would say, don't be fooled by overzealous, nonsensical, over-reaching, local, made-up, (either fetishistic, or otherwise control freak) rules, which stand in the way of Wikipedia doing its job, part of which is getting the names of things right. Proper names exist outside of Wikipedia - the fact that they are not allowed to persist inside Wikipedia is an aberration, unsupported by widely practiced standards of other encyclopedias and scholarly works. --Lexein (talk) 04:35, 21 September 2011 (UTC)

OK, like "telephony" was originally the title of an implementation. Let's get over this obsessive upcasing of anything that is vaguely technical, and restrict upcasing, as required by our site policies and guidelines, to items that are unambiguously proper names. And no, what someone does "outside" doesn't dictate what we do, or WP would be an unholy mess. Every publication, including ours, has house rules. Tony (talk) 06:42, 21 September 2011 (UTC)

We are not talking about an implementation here, but about a single, specific solution. I'm not sure how you can compare this to telephony, which is the general concept of communicating in voice over distance. Anyway, as I said above it depends on whether there is a general meaning for "Internet relay chat" or whether the meaning applies only for the specific, standardized protocol. I am uncertain about the more general usage, which is why I just commented, but I do think that if the name is wished to be lower-cased this more general usage must be proven via references. Nageh (talk) 07:40, 21 September 2011 (UTC)

Tony: "Vaguely technical" - that's blinkered: deliberately not seeing or understanding, sorry. Where is this bias against getting the names of things right coming from? House rules should have some grounding in reality, since the aim of Wikipedia is after all, to be an encyclopedia, comparable, presumably favorably, with other encyclopedias.

Nageh: There is no general usage with the capitalization "Internet relay chat", except by mistake, or ignorance. The "United states of America", or "North america" : both are just wrong, against the proper name and proper usage. --Lexein (talk) 07:49, 21 September 2011 (UTC)

Geographical names are always upcased. Technical terms not so unless there's a particular reason. Perhaps we could more usefully negotiate what those reasons are, to guide future discussions; there's quite a big grey area in telecom and computing, it seems to me, and non- and semi-experts (as well as experts) need to be in greater agreement about up and down. Tony (talk) 08:38, 21 September 2011 (UTC)

As well as distinguish noun from verb. "Internet Relay Chat" is a noun, "to chat" is a verb. I'm fine with lowercaseing the verb, of course. But nouns always get the proper name form, if there is one, and here, there very specifically, historically, is one. "Bell Telephone Co.", noun, "to telephone", verb, "to use a telephone", lowercase noun. "Transmission Control Protocol", noun, upper. "Per section IIa of the protocol", lower, except in the case of matters of state or medicine. These are all fine rules which work everywhere, except, apparently, here. --Lexein (talk) 09:25, 21 September 2011 (UTC)

There's always a possibility that I could be convinced to change my position but I have better things to do than participate in disrespectful discussion. Jojalozzo 15:10, 21 September 2011 (UTC)

Well, that's too bad. My position is that there is a fundamental disrespect for authors and creators of works, and the readers of Wikipedia, built into the current MOS, against calling things by their proper names; it is against this that I reason, argue, and rail. You should see that as a valid position, rather than sniffing and refusing to contribute to the discussion. --Lexein (talk) 01:42, 22 September 2011 (UTC)

• Oppose This is a proper noun. - MrOllie (talk) 15:15, 21 September 2011 (UTC)
• Oppose Internet Relay Chat as per rfc1459 and capitalisation example MAVEN. --Hm2k (talk) 16:04, 21 September 2011 (UTC)
• Oppose "Internet Relay Chat" is this Internet protocol's proper name. This rename request would be like renaming World Wide Web to "World wide web", File Transfer Protocol to "File transfer protocol", Domain Name System to "Domain name system", Simple Mail Transfer Protocol to "Simple mail transfer protocol", etc. In short, it ain't gonna happen. --Tothwolf (talk) 02:17, 22 September 2011 (UTC)
• Oppose This is the specific IRC protocol, as defined by a rfc. This not about the generic concept of "chat that happens to be relayed over internet". --Enric Naval (talk) 21:27, 26 September 2011 (UTC)

The above discussion is preserved as an archive of the proposal. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

IRC-Galleria?

An editor has suggested a link to the IRC-Galleria article ("the largest social networking website in Finland") be included in the See Also section of this article. I do not believe the link belongs here, any more than does, say, a link to 4chan or Facebook. Perhaps it should be listed on the IRC dab page (and it may be; I've not looked), but I do not believe it belongs in this article.

I reverted the addition of the link for this reason. Does anyone have any thoughts on the matter? — UncleBubba ^{( T @ C )} 03:20, 7 December 2011 (UTC)

Over-detailed?

An editor (the same one who tried to add IRC-Galleria to the EL list) has added an "over detailed" and "condense" tag to the beginning of this article. I disagree, but am not uninvolved, so I would like other opinions.

This page attempts to cover a technical topic and has roughly the same number of sections and level of detail as do the HTTP and FTP pages. I don't see that as problematic.

Anyone have any thoughts? — UncleBubba ^{( T @ C )} 14:28, 9 December 2011 (UTC)

I have removed the templates for the reasons stated above. — UncleBubba ^{( T @ C )} 14:34, 2 January 2012 (UTC)