I just posted a new version of rto-consider (-16). The document, diffs, etc. can be found here: https://datatracker.ietf.org/doc/draft-ietf-tcpm-rto-consider/ I fixed a few nits that had been pointed out throughout the document. The main work in this rev is in the first two paragraphs of the intro. I honed these based on the discussion with Stewart and Gorry. Well, "honed" might ultimately look like "rewrote" even if that wasn't the process or exactly my intention. :) I tried to pull some of the notions from later in the document up here to the front of the document, as requested. In particular, I brought forward the notion of this document as a default and not as a one-size-fits-all. I hope this addresses the concerns to a large extent. I also want to make a meta point here, which is clearly about this document, but also I think much more widely applicable. I continue to believe that the Internet has grown to what it is in no small part because at many crucial points we chose generality over optimality. I do not believe we should shun generality---either in terms of protocol development or the lessons we have learned. That doesn't mean we cannot acknowledge places where more pointed solutions may be appropriate. It doesn't mean we can't accommodate different and non-general approaches in things like guidelines and requirements. But, I think it'd be a pity if we couldn't write general documents because there are cases where they will be suboptimal or not applicable. There are always such cases. And, if that is the gate, we're going to design a network we won't much care for, I think. IMHO. FWIW. allman
Attachment:
signature.asc
Description: OpenPGP digital signature
-- last-call mailing list last-call@xxxxxxxx https://www.ietf.org/mailman/listinfo/last-call
