On Fri, Mar 3, 2023 at 4:33 PM David Schinazi <dschinazi.ietf@xxxxxxxxx> wrote:
Hi Vijay, and thank you for your review!The distinction between MUST and SHALL in this document is intentional. We use MUST to indicate "if X happens, then you MUST do Y" whereas we use SHALL to indicate "The procedure is as follows: step 1 you SHALL do Z". Based on previous RFCs published by this WG and on guidance from the IESG, the choice between MUST and SHALL lies with the editors. Therefore, I respectfully disagree with your advice and will leave the text as-is.
Dear David: Thank you for your time on my Gen-ART review. Of course, the review serves in an advisory capacity, so your editorial discretion holds, and I am certainly not trying to upend it, or the precedent set by previous RFCs published in the WG.
However, there is a vast legion of implementers who do not participate in the IETF, but still implement the standards produced by the organization. The distinction you make above is lost to that audience, who as I said will probably resonate with a MUST rather than a SHALL. (Certainly, as a long-time IETF participant, this distinction was lost on me as well when I was reading your draft. It certainly is a distinction that RFC 2119 does not make, at least in my reading of it.) But nonetheless, this is a stylistic editorial choice driven by past precedent in the WG; although I do think that if you wanted to put the rationale you provide above in the I-D, that may not be a bad idea. After all, we do produce these documents so that they are understood unambiguously by as wide a cross-section of implementers as possible.
Cheers,
- vijay
-- last-call mailing list last-call@xxxxxxxx https://www.ietf.org/mailman/listinfo/last-call
