Re: Proposed IESG Statement on the use of the “Updates” header

[Ted Lemon <mellon@xxxxxxxxx>]

On Sep 12, 2018, at 2:18 PM, Adam Roach <adam@xxxxxxxxxxx> wrote:

I still don't follow. If the abstract does not contain enough information to let someone know whether they want to read the rest of the RFC, then what purpose *does* it serve? I note that many (non-IETF) protocol specifications are published without an abstract at all. If ours doesn't serve any purpose, then perhaps it's time we discussed whether RFCs need them at all [1].

Abstracts are useful in precisely the way you suggest; the question is whether the abstract has to enumerate what the document updates.

This may not be necessary if the reader would be reading the document anyway because of the contents of the document.   So whether the abstract has to say specifically what is update is a judgment call.   Personally, I prefer to keep abstracts as short as possible—they should definitely say enough that I can decide whether I need to read them, but if they say more than that, then reading the abstract becomes enough work that it might as well just be an introduction (and indeed we often see IETF documents where the abstract and the introduction are largely the same text).