RE: Straw poll: what options did you review?

> As a note on this straw poll, I think that the question of whether you
> have reviewed all the options in the protocol is not very relevant.
> Virtually nobody has expressed an interest in implementing all the
> options (at least, not in their first release).
>
> The relevant number is whether a sufficient number of people have
> reviewed each option (which in fairness to Jim, was his question
> number two).  And since the options are cleanly segregated into
> separate sections, I'd be happy to "defer" any option that does not
> have sufficient review/support.  My criteria for sufficient support is
> that there are two different implementors planning on using it.

Well, ideally there will have been a number of reviews of the entire
document, but a high average number of reviews/option is also a good sign.

I'm not interested in defering options that haven't been sufficiently
reviewed, I just want to make sure they have been thoroughly reviewed. If
lack of reviews of one or more options is the critical path, I suspect these
sections will get reviewed quickly.

I really want to see this document progress as quickly as possible, but I
also need to know that it has been reviewed by a couple different pairs of
eyes.  It's much cheaper to fix errors now, by several orders of magnitude,
than after the spec. has been implemented N times.

- Jim

Received on Wednesday, 7 February 2001 20:35:48 UTC