On 11/7/19 8:31 PM, Brian E Carpenter wrote:
Anyway: my new experiment would be one that the IESG could decide to start tomorrow. It's simply that the IESG would only ever issue one form of DISCUSS ballot, which would look like this: Pat Areadirector has entered the following ballot position for draft-ietf-somewg-somedraft-99: Discuss .... ---------------------------------------------------------------------- DISCUSS: ---------------------------------------------------------------------- There are still open issues from the following reviews: <links to reviews> In other words, the IESG simply busy-waits until all review issues have been resolved, rather than finding and fixing the issues personally. (If an AD wishes to post a review for a given draft, that would be a personal choice, not part of the IESG workload.) This is a procedural change, and would not prevent a substantive DISCUSS in unusual circumstances.
To me the problems seem like: (a) getting sufficient reviews done, and (b) establishing review criteria well in advance, not only for the reviewers but also as feed-forward to the WG and document editors. (not that reviewers couldn't identify issues outside of those criteria, but having a somewhat-shared view of those criteria would make it easier to compare different reviews for example)
I keep looking for ways to identify these criteria earlier in the process. Not that there would never be last-minute showstoppers - we're always learning new things. But if we could reduce the number of late surprises that would make everything easier IMO.
Keith
