Re: [rfc-i] "Obsoleting" a perfectly valid document

Hi Дилян,

sorry for not giving more context.  The text I don’t like actually now is in https://tools.ietf.org/html/draft-ietf-cbor-7049bis-06, twice, in the abstract and at the end of the introductory text of Section 1.

I do understand the language we use inside the IETF for document revisions, and I agree that the sentence we wrote is clear for readers that know that special language.  Actually, giving that information is necessary for the reason you give:  So people know which documents they no longer need to read.

However, for someone who isn’t as familiar with RFCs, there is a significant danger of misunderstanding the term “obsoletes”: 

> obsolete | ˌɑbsəˈlit | 
> verb [with object] chiefly US 
> cause (a product or idea) to be or become obsolete by replacing it with something new: “we’re trying to stimulate the business by obsoleting last year’s designs”. 
[cited from New Oxford American Dictionary]

But the idea or the product are not obsolete, only the specification has been revised after 5 years of experience with people using it.  People use RFC numbers both to refer to the actual RFC and to the concept defined in there (as in “an RFC822 message”, of course usually with the number of an obsoleted RFC…).

My message was a request for help writing this up in a better way, but also a repeat of my ceterum censeo that some of the language we use to describe our processes is incomprehensible to outsiders.

Paul Hoffman has fixed up my proposed text a bit (https://github.com/cbor-wg/CBORbis/pull/89), and with the support here I think this can go in.  Still, I would like to hear about ways other people have handled similar situations — it is quite tedious to extract this kind of information from some 1140 RFCs that obsolete others.

(Just to give one example: RFC 2822, which obsoleted RFC 822 and was then obsoleted by RFC 5322, never says so except in the head of the front page.  It also goes ahead and defines a whole section of “obsolete syntax”, which retains some no longer desirable parts of RFC 822 only to keep some measure of backward compatibility.  None of this changed in RFC 5322.  Oh, and RFC 822 obsoletes RFC 733, but uses that term again only in its front page head, and uses “revises” in the text to describe the not insignificant technical changes that were made.)

Grüße, Carsten

_______________________________________________
rfc-interest mailing list
rfc-interest@rfc-editor.org
https://www.rfc-editor.org/mailman/listinfo/rfc-interest