Make obsolete documents more obviously obsolete

[martinthomson]

Description

See https://mailarchive.ietf.org/arch/msg/last-call/P6HaBJHlv1lvL6dUJbXCmA7Bb3o for context.

Consider the obsolete RFC 793 and its presentation on datatracker: https://datatracker.ietf.org/doc/html/rfc793 and https://datatracker.ietf.org/doc/rfc793/

Both of these show the document as an "Internet Standard" in the most prominent (and green-colored) position. You have to dig a little more to realize that this is obsolete.

Though the mail thread contemplates formally adding an "obsolete" status for documents, there are UX improvements that could make that condition more prominent, so that people spend less time thinking that an obsolete document is current and valid.

I'm not going to do UX design here, as much as I'd like to, but I think that some small changes to the presentations would help a great deal.

Code of Conduct

• I agree to follow the IETF's Code of Conduct

[becarpenter]

I support such an enhancement, and without impinging too much on UX myself, simply displaying "OBSOLETE INTERNET STANDARD" or "OBSOLETE PROPOSED STANDARD" etc. wherever appropriate would help a lot. "OBSOLETE INFORMATIONAL" wouldn't harm, either.

[cabo]

On 18. Dec 2024, at 03:28, Brian E Carpenter ***@***.***> wrote: I support such an enhancement, and without impinging too much on UX myself, simply displaying "OBSOLETE INTERNET STANDARD" or "OBSOLETE PROPOSED STANDARD" etc. wherever appropriate would help a lot. "OBSOLETE INFORMATIONAL" wouldn't harm, either.

I don’t think TCP is obsolete. Calling superseded documents “obsolete” is not very bright. (I know that this is our official terminology.) We don’t need to rub this mistake into the eyes of unsuspecting users stumbling over RFC 793. In a Web UI intended for the wider public, please do use English words such as superseded in place of terms that only have meaning to people who have been in the IETF for 5+ years and are actively misleading for everyone else. Grüße, Carsten

[martinthomson]

Fine, "OUTDATED VERSION (see RFC $REPLACEMENT)"

Though if I were to be pedantic (it's not a competition) the document is indeed obsolete, even if the technology isn't.

[cabo]

Fine, "OUTDATED VERSION (see RFC $REPLACEMENT)"

WFM (although it perpetuates the use of my pet non-term, "version").

Though if I were to be pedantic (it's not a competition) the document is indeed obsolete, even if the technology isn't.

Yes. The problem is this linguistic device called "pars pro toto", in where we use a part of something (Washington) to refer to the whole thing (US of A). "RFC 793 is obsolete" means "the document RFC 793 is obsolete" as much as it means "TCP the technology, which is described in RFC 793 so we use this as its pars pro toto handle, is obsolete". Clearly, RFC 5246 is obsolete in both ways, so this interpretation wouldn't be entirely mistaken.

[OR13]

Proposal:

Badging should be used consistently, knowing that something is obsoleted is also possibly more important than knowing that it has errata, the order here should probably be errata second and not first... I would also make errata yellow and obsoleted red.

[paulwouters]

I am in favour of clearly marking the obsoleteness beyond just the one coloured tag.

[becarpenter]

Back to @cabo's comment, REPLACED INTERNET STANDARD seems better overall. There is indeed a semantic difference between replacing a standard and obsoleting it. And if you imagine "Replaced" where @or18 has "Obsoleted" it works fine.