Talk:Request for Comments

From Wikipedia, the free encyclopedia
Jump to: navigation, search
High traffic

On 2009-04-08, Request for Comments was linked from Slashdot, a high-traffic website. (See visitor traffic)

Broken RFC Links?[edit]

Seems all of the RFC's that are linked to are returning 404's, with the amount listed here, and elsewhere, would some regex wizard be able to relink them correctly? —Preceding unsigned comment added by (talk) 05:27, 13 January 2009 (UTC)


I found this in the German Wikipedia. Maybe someone could add a description. Don't know how to describe it in english. I hope it's not counted as an advertisment. I'm not a supportor or afflite of this (yet), but it sounds interesting to me. The link is: —Preceding unsigned comment added by (talk) 21:30, 29 November 2007 (UTC)

Important RFCs[edit]

It would be really nice if someone went through that list of "important" RFCs and added some information saying why each is important. If no-one does it soon, I'll do it... -- Timwi 22:59 20 Jun 2003 (UTC)

I presume by important you mean promenent and/or widely used ?
I am referring to the list with the heading "Here is the list of the most important RFCs:". -- Timwi 15:14 23 Jun 2003 (UTC)

Separate RFC articles merged here[edit]

Lots of RFC articles have been merged into this one. Angela did this but takes no blame for it. From VfD:

  • It appears as though all but the first of these has not been edited in quite a while (the first I made a complete sentence before deciding to list them here), if ever, and now are listed on the ancient pages list. They consist entirely of an external link and what appears to be a citation of the external link. I can only assume the topics are encyclopedic, but Wikipedia is not a depository of links. Tuf-Kat 03:23, Sep 24, 2003 (UTC)
  • Delete.Vancouverguy 03:30, 24 Sep 2003 (UTC)
  • Merge the lot of them into RFC. -- Cimon Avaro on a pogo-stick 15:56, Sep 24, 2003 (UTC)
  • Vote againt. They should be merged or redirected, which is good enough. -- Taku 21:59, 26 Sep 2003 (UTC)

Is there any particular reason you have multiple "External Link" headers? Also, may I suggest that we have a separate "Links to RFC's" section for the links to actual RFC's, and not lump them in with the rest of the random external links? Finally, to make it easier to edit, we might want to have several sections of RFC links (titled something more imaginative than "Links to RFC's in the 1-1000 Range" :-). Noel 15:22, 1 Oct 2003 (UTC)

There is a reason - I'm lazy. :) There were millions of the silly things so I just pasted them all in and redirected the originals here without actually trying to reformat them or make the article make sense. If you want to sort it out Noel, I would be very grateful. If not I will fix it at some point, but not today. Angela 19:49, 1 Oct 2003 (UTC)
Well, I've taken a crack at it. It's a bizarre selection there, some of the more important ones are missing. While I was at it, I switched to a common entry format. If someone's really energetic, they can go fix all the [RFC-foo] forwards so they link to the relevant sub-section here.
In retrospect, it might have been better to leave the individual pages (since if we add all the important ones, this page is going to get really long), and add a sentence or two to each saying what each one covers. Oh well, that's what I get for not reading VfD. Noel 21:23, 1 Oct 2003 (UTC)
You could always revert them all. Actually, as it was my fault, I should probably be the one responsible for undoing it. I don't mind if you decide that is the best option. Anyhow, you've done an excellent job on it so far.
Regarding the redirects, you can't redirect to specific sections yet but DanKeshet has submitted it as a feature request at Sourceforge. Angela 23:04, Oct 1, 2003 (UTC)
Nah, it's not worth all the trouble to undo it. About the redirects, I thought we could pick out subsections. Maybe it's only in URL references to the Wiki, though. Noel 03:03, 2 Oct 2003 (UTC)
Ok, if you're sure. Do you mean links or redirects? You can definitely link to subsections. I thought you meant #redirect[[Request for comment#foo]] - that wouldn't work. Angela 03:16, Oct 2, 2003 (UTC)
Well, I'm sure I don't want either of us wasting time putting them back! :-) Yes, I did originally mean redirect to subsections - I didn't know that didn't work. Noel 12:37, 2 Oct 2003 (UTC)

Wikipedia:Request for comments link[edit]

I don't understand why people keep removing the link to Wikipedia:Request for comments. It is very apropos and useful here. By comparison, VfD redirects to Wikipedia:Votes for deletion, Rfa redirects to Wikipedia:Requests for adminship, and so on; is it that unreasonable that RFC and Request for Comments should provide at least a link to the W:RfC page? Someone could well be seeking that page here. Also, cf. VFD. -- VV 23:50, 4 Apr 2004 (UTC)

Etymology of "Foo"[edit]

I find it somewhat amusing that RFC 3092 (Etymology of "Foo") is listed among the most important RFCs... :) Fredrik 20:34, 7 Apr 2004 (UTC)

Three-digit RFC numbers[edit]

Are three-digit RFCs supposed to be listed as RFC 0123 or as RFC 123? Both exist on the page. It's fine to do it either way, but some standard should be settled on. I vote for the second way, dropping the leading zero. Grendelkhan 03:05, 2004 Apr 20 (UTC)

Another vote for dropping the leading zero. There are no leading zeros in the RFCs themselves. —Bkell 06:23, 21 Apr 2004 (UTC)
++dropping Guaka 22:25, 23 Apr 2004 (UTC)

Joke RFC?[edit]

RFC 439 seems like a joke to me. But it dates January 21... Guaka 22:25, 23 Apr 2004 (UTC)

I don't think it's a joke; it appears to be an actual conversation between two artificial intelligence systems. Granted, it's entirely for entertainment value, but there's no restrictions on what RFCs can and cannot contain. —Bkell 05:56, 25 Apr 2004 (UTC)
Yes, it is real - the two programs (Parry and the Doctor) really existed, and this is (AFAIK) what they really "said" to each other when hooked together. (I seem to recall that this was actually printed in Datamation, back when it was done.) Things weren't as formal that far back in time, and so you could get this kind of thing on dates other than April 1, something that might not happen now. Noel (talk) 02:23, 6 Dec 2004 (UTC)

Added table and wikified[edit]

I wikified the RFC links (i.e., prefer RFC 1234 = RFC 1234 to [ RFC 1234] = RFC 1234) and added all the information to a table, which I feel is the proper way to present data repeated in a regular format. 02:09, Sep 5, 2004 (UTC)

Wouldn't it be better if RFC was wikified to Request for Comments, and the RFC number to the RFC text at ? So readers could read general info on RFCs in the article. Apokrif 21:47, 4 May 2006 (UTC)

Origin of the name[edit]

Perhaps someone should write why it's called Request for Comments, I have no idea, fo me it's a strange way to define a standard to calle it rfc. -- AzaToth 11:58, 3 Nov 2004 (UTC)

History, basically. Back in the days of RFC 1, it was just some notes being passed back and forth among a small group of researchers. It has gradually become a lot more formal, but the name remains the same, mostly because it would be too confusing to change it. There are additional series now, STD and BCP, but RFC remains the one everyone knows. Noel (talk) 23:20, 15 Nov 2004 (UTC)
I've recently edited the article, I will add this in now (I imagine other people might want to know too). Noel (talk) 02:23, 6 Dec 2004 (UTC)

Proposal to split the page[edit]

I think this is unwieldy, and hard to use in it current large size. I propose that we split it up, and create List of the most important RFCs and List of IETF RFCs as separate articles. Noel (talk) 02:48, 6 Dec 2004 (UTC)

Or something. It takes up half the bloody page, and it's not terribly useful. We even have a smaller list higher up on the page. Exactly what use does the enormous RFC list serve? What does it do? If someone wants to know what RFC defines IMAP, they can go to IMAP and see it there, where it's apropos. This is inflating the page for no real purpose. I don't think the page should be split, I think it should be pared down. grendel|khan 17:14, 2005 May 6 (UTC)
Besides being big, it is misleading. Wikipedia should avoid labeling the status of documents (e.g. Proposed Standard) because it changes over time, and this is not a likely place that it will be updated when necessary. Just refer folks to the rfc-editor pages for the daily updates of current status. Adding the definitions of the status options would be a good addition to this page, though. --NealMcB July 3, 2005 04:16 (UTC)
Agreed. And it's a mounting problem—people continue to create new lists and expand old ones. Where does it end? Surely there are better ways to make this article a valuable resource. – Ringbang 15:04, 13 November 2005 (UTC)

Random RFCs?[edit]

Why a list of "random RFCs"? Maybe "Key RFCs", or "Importent RFCs"? ~ mlk 03:19, 30 Dec 2004 (UTC) ~

Imporant RFCs (again)[edit]

The list of "important RFC's" is still really long, and now that each entry links to the actual RFC, has much duplication with the later lengthy list of RFC's at the end of the article. I don't have the energy to do much about this at this point - I just spent a couple of hours going through the list of "important RFC's" and trying to weed out the obsolete/obscure ones, and making sure all the important ones were there. Noel 18:36, 5 Mar 2004 (UTC)

Blast! 821 and 822 are listed as "obsoleted by 2821 and 2822", so I'm amazed to find that their status is "STANDARD" and not "HISTORIC". In general, I didn't list any RFC as "important" if it was listed as "obsoleted by RFC xxx". (In fact, I started off by adding those two, and later took them out when I found them listed as "obsoleted by".) Must be a temporary glitch in the system. Noel 03:27, 20 Mar 2004 (UTC)
Not a glitch, except in the sense that the IETF process doesn't quite make sense when a new version of a Standard protocol is introduced at a Proposed Standard level. Blame it on RFC 2026. Alvestrand 01:30, 26 December 2005 (UTC)

Oldest Public File on the Internet[edit]

I once tried to find the oldest public file on the internet and I came to the conclusion it was RFC 1 (filename '1') which I found on a university ftp with the timestamp of 7 April 1969. I can't seem to find this file anymore, does anyone know of it?

I just disproved my own theory when I read this at 'Crocker wrote the RFC late at night in his bathroom [on a piece of paper], so as to avoid awakening his roommates.' Looks like the file I found had a bogus filestamp.
It isn't a bogus timestamp, it's the date that it was written. It says the date that it was put into machine readable form near the bottom of the file. --P4ch3c0 00:23, 30 March 2006 (UTC)

Lists of RFCs removed[edit]

Yes, I removed all of the lists. I know there will be mixed reactions, but here is the rationale:

  1. The lists contravene Wikipedia policy. See Wikipedia is not a mirror or a repository of links, images, or media files.
  2. The "External links" section contains several links to sites that categorize RFCs by subject, classification (BCP, Internet standard, etc.), and import. Also, some of the lists were just old reproductions of updated lists mirrored elsewhere.
  3. Wikipedia articles about technologies defined in RFCs already provide links to their respective RFCs; for centralized repositories, there are better resources (see also point 1).
  4. Linking to any RFCs in a list or tabular format invites others to expand the lists—and we've already seen where that leads us.
  5. If a particular RFC was pivotal in the history of the IETF and the concept of RFCs (e.g., RFC 1), it can be mentioned in the text. On submission, an inline reference in the form of "RFC #" is automatically parsed into a link to the official document.

Ringbang 14:14, 14 November 2005 (UTC)


FYI: Maybe we could arrange a better link for RFC 4321 anywhere in plain text, see MediaWiki talk:Rfcurl. -- Omniplex 16:40, 16 April 2006 (UTC)

Being an IMP?[edit]

The History section includes the phrase "because UCLA was one of the first Interface Message Processors (IMPs) on ARPANET.". Is that really correct usage of the term Interface Message Processor? It's my understanding that an IMP was the interface to the network, like a NIC is on a LAN. How about "because UCLA had one of the first IMPs on ARPANET"? JöG 20:27, 30 September 2006 (UTC)

This article may need semi-protection[edit]

This article for some reason has been repeatedly vandalized and significant content was lost since May, until I noticed what was going on this afternoon. If the pattern continues, semi-protection may be necessary. I am flagging the issue so the admins can monitor this article and act accordingly. --Coolcaesar 06:55, 27 July 2007 (UTC)

IETF RFC handling in Wikipedia; best way to cite[edit]

RFC references are automagically linked in Wikipedia, but I haven't run across anything that documents this. I'm absolutely certain I could if I dug harder, but that's kind of my point: shouldn't there be at least a passing mention of it here, presumably as a hatnote? I know this would clutter the article a bit, so alternatively, maybe it could be mentioned in a hatnote on Wikipedia:Request for Comment? If not, where should I be looking?

The reason I really bring this up, other than the obvious "feature documentation should be easy to find" point, is a related topic: what's the best way to cite an RFC?

I'm currently editing an article that has such a cite, which currently uses the {{cite web}} template. The problem, as I see it, is that this template requires a URL. Hardwiring a URL in this fashion seems to me to work against the WP "automagic RFC link" feature, since the automagic link is presumably to what is felt by consensus to be a canonical source, which may change some time in the future. If RFC URLs are also hardwired in {{cite web}} links scattered throughtout articles, this defeats one of the purposes of having the automagic links.

I think the proper solution would be to have a new dedicated {{cite rfc}} template that would require a serial number field instead of a URL field, and use the same routines as the automagic to generate the appropriate link. It could then also have only the fields appropriate to an RFC, rather than the more generic {{cite web}} fields.

Or maybe the right answer is "don't cite RFCs at all; let the automagic stand in place of a cite." This doesn't seem entirely sufficient to me.

This is probably the wrong place to discuss this, but I'm not sure where the right place would be...--NapoliRoma (talk) 20:51, 13 January 2008 (UTC)

  • I agree, a bunch of stuff should be done with IETF RFCs in the wikipedia. Addressing your points:
    • The only mention of the special handling of RFC#### that I've found is in Wikipedia:How_to_edit_a_page#Links_and_URLs, and that could be improved
    • There is a Wikipedia:ISBN page for the similar handling of ISBN####, and I think that maybe a short Wikipedia:IETF RFC page should be created, with a hat-note on both Wikipedia:RFC and this page.
    • I think it would be useful to have a working {{cite rfc}}, especially one that could give the section numbers. There are many times when you don't really want to say "blah blah as defined in RFC 2821 section 2.3.10", but rather "blah blah{{cite RFC 2821|section=2.3.10}} and have the actual footnote not use the built-in RFC linking, but rather a direct link to the URL such as
    • A little over a year ago, I went through and tried to standardize wikipedia to use the built-in RFC linker when at all possible. See my User:Wrs1864 page for a list of websites that people frequently link to instead of using the built in system.
    • I also, at the same time, got rid of most of the RFC #### articles and redirects since I frequently found people linking to them via [[RFC 1234]] instead of just typing RFC 1234. See Wikipedia:Redirects_for_discussion/Log/2006_December_21#RFC_1_.E2.86.92_Request_for_Comments for a lot of discussion
I have not had much time lately to do more work on this area. If anyone could help out, I think it would improve the wikipedia. Please don't count on me doing much more in the near future. Wrs1864 (talk) 20:03, 17 January 2008 (UTC)

Along the lines of the URN, shouldn't a citation template (or anything like the current auto-linking system) ideally handle all the various types in the series (RFC, STD, BCP, etc.)? And how should the multiple identities of a single document be handled? If an article mentions "RFC 5000", should the citation automatically mention that that document is also called "STD 1", or would the complexities of such a feature make it too difficult to implement? What should the behavior be for Standards that are in transition? Example: RFCs 821 used to be an Internet Standard document, part of STD 10, but it's been superseded by a mere Proposed Standard, RFC 2821. See Internet Protocol Standards. It's not completely obvious how to handle such a situation if it's handled automatically, but if it's left to the editor, different editors will handle it in very different ways. What would the best automatic or template behavior be?

Stephan Leeds (talk) 12:21, 27 May 2008 (UTC)

For what is worth, I recently started looking into writing a {{cite rfc}} template, but someone does it first, that would be great. As far as your questions go, yes, I think template should handle all the series, although this could be a problem since the website doesn't handle the fyi and std series, just the rfc and bcp series. I can not see an easy way of figuring out inside a template that an RFC happens to also be a STD/FYI/BCP, so I wasn't planning on doing that. I was going to make the template as simple as I could and wasn't going to deal with problems like you mentioned with STD 10. In my searching of RFC references, I found that to be an extremely rare situation so leaving it to the editor shouldn't be a large problem. Wrs1864 (talk) 12:57, 27 May 2008 (UTC)
The URL for the automatical links to RFCs is defined in MediaWiki:rfcurl, there is a talk page with the history. Meanwhile the IETF Tools Server URLs are the MediaWiki default, and I'm sure that this is documented in a Meta help page, but I'm too lazy to look where. You can get the same effect as RFC 2345 by writing {{int:rfcurl|2345}}, demo: // The ISBN magic is more convoluted, there is a MediaWiki page like {{int:rfcurl}} pointing to a project page, default Project:Book sources, and the ISBN magic uses the designated book sources page as a kind of template replacing a magic word by the given ISBN (not what we know as "template" and "magic word", it's a specific magic for ISBNs). I don't see the point of using urn:ietf:rfc:4408 (or similar) without a resolver. The IETF considers to create an ISSN for the whole RFC series, maybe that helps (?) -- (talk) 18:36, 28 May 2008 (UTC)

URI for RFCs[edit]

How about a section covering the URI for RFCs? For example, urn:ietf:rfc:2141 .

The section could cover the theory and potential and actual use. Potential use could be a browser plugin/extension which would enable one to turn <a href="urn:ietf:rfc:2141">such an link</a> into a link to the RFC at the user's favorite RFC mirror. Actual use of the URN on the web right now is non-existant as far as I know. (talk) 16:09, 18 April 2008 (UTC)

Write about RFC people[edit]

In early years of networking only relatively small number of people was engaged in creating standards and thus had enormous influence on modern networking. It would be nice to mention them here. —Preceding unsigned comment added by (talk) 16:55, 11 October 2009 (UTC)

Not sure it belongs in this article - but just as an exercise, it could be interesting to pair up the list of people with many RFCs with the list of people on Wikipedia and see who's in both places.... while # of RFCs does not necessarily correlate to influence, some of the people with many RFCs are influential. --Alvestrand (talk) 05:45, 12 October 2009 (UTC)
In fact this seemed like so much fun that I started a list. See User:Alvestrand/RFC authors for status. --Alvestrand (talk) 06:21, 12 October 2009 (UTC)
There is also RFC 1336:
Malkin, Gary (May 1992). Who's Who in the Internet: Biographies of IAB, IESG and IRSG Members. IETF. RFC 1336. Retrieved October 21, 2009. 
--Tothwolf (talk) 14:56, 21 October 2009 (UTC)

Explanation of change[edit]

The article should explain why it is that Requests for Comments changed from requests for comments into final standards. — Preceding unsigned comment added by (talk) 07:45, 15 May 2015 (UTC)


Are the results of an RFC just a vote count, or do the !votes have to invoke Wikipedia !rules and Wikipedia !law? Is calling something trivia a valid RFC comment that counts as a !vote? --Richard Arthur Norton (1958- ) (talk) 15:53, 24 November 2015 (UTC)