Re: [Ltru] RFC4645 to Historic status

[Doug Ewell <[e-mail]>]

Timothy Mcsweeney wrote:

> I think changing the link was a good idea because it was the source of
> confusion.  After more thought, it would be better(and more efficient
> for all) if it were labled Historic or removed all together so the
> next person who may come across the document won't be forced to read
> all the way up to the link that tells them they just wasted thier
> time, now possibly on purpose.

Everyone has his own source of confusion in this discussion, and this is mine.

This all seems to have started with Tim's review of the OpenID Connect Core 1.0 draft (https://openid.net/specs/openid-connect-core-1_0-31.html), which provides for specifying languages and locales, and as such contains the following normative references, with the HTML link listed below each (because this is a plain-text email):

[RFC5646] Phillips, A., Ed. and M. Davis, Ed., "Tags for Identifying Languages," BCP 47, RFC 5646, DOI 10.17487/RFC5646, September 2009.
https://www.rfc-editor.org/info/rfc5646

[IANA.Language] IANA, "Language Subtag Registry."
https://www.iana.org/assignments/language-subtag-registry

Whenever any reviewer has questions about how a process or protocol described in a draft uses some standard or other external document, then not only the right thing, but the *obvious* thing to do is to look at the NORMATIVE references in the draft. That is why IETF drafts, and most other reasonable specification, must include such references.

How Tim found himself at RFC 4645 and the IANA "Protocol Registries" page, when the draft under review contains perfectly good references to the RFC and registry that ARE relevant, is the part that escapes me.

I agree, as I said before, that someone randomly stumbling across the IANA Protocol Registries page could easily be confused by its intermixing of current and obsolete language-tag links. That does need to be cleaned up, possibly by clearly separating out the old references, or possibly by removing them. (RFC 3066 was replaced 17 years ago, after all.) But there is NO reason why anyone reviewing a draft that refers explicitly to BCP 47 and the LSR should find themselves randomly stumbling along this path, and then claim confusion.

I have no idea what Tim's comment about being "forced to read all the way up to the link" is about. The OpenID document, in HTML format as he reviewed it, only requires one to click wherever it says [BCP47] or [RFC5646] or [IANA.Language] in order to find the right reference. And if one does find oneself at RFC 4645, for whatever reason, that document also makes its scope abundantly clear at the outset.

I can’t even imagine what "now possibly on purpose" refers to.

--
Doug Ewell, CC, ALB | Lakewood, CO, US | ewellic.org