[rfc-i] Categories of references

Carsten Bormann cabo at tzi.org
Fri Oct 3 13:12:11 PDT 2014


Thank you all for the temperature readings on this.

No, I didn’t want to generate more (or even change existing) process, I apologize if the summary at the end sounded that way (it simply echoes the way I think about the existing categories, but you don’t have to agree with that).

I was just interested in an additional tool for those RFC authors who care about the workload of the specification readers.  Yes, that tool is more useful in the tomes with 50+ references we sometimes generate than in a 5-pager.

Heather has pointed out what I think is the way forward: enabling an individual RFC author to give a substructure to X.2 “Informative References”.
The somewhat lame reason I didn’t fully go for that initially is that I simply don’t know a good title for X.2.1, i.e., the informative references I actually want the spec readers to take note of (“Recommended Reading”? ouch).  (X.2.2 can then simply be called “Historical References”, that one is easy.)
The good thing is that we don’t necessarily have to come up with a one size fits all solution here.

So the next thing I’ll do is check the tool support for this and try it out in one of the next I-Ds I’ll co-author.

Grüße, Carsten

PS.: Martin: I don’t think the way we are using references in the text today typically is that indicative of the purpose of an informative reference.  My example was deliberately overstated...  Giving the reader a few more hints is a useful tool here that an RFC author may or may not want to pick up.

PPS.: Oh, and many things are wrong with the way we are using the distinction between normative and informative references today.  But fixing that would take so much effort (real process changes and all that) — and most likely simply end up replacing the old problems with new problems — that I don’t think investing time there is worth it.



More information about the rfc-interest mailing list