Andrew,
I guess I'm sceptical that this problem can really be resolved by more process rules.
I think we need to separate a few different problems and what the remedy for them might be.
1) The ball is simply dropped. Do we have this problem? Yes we do, at least for some number of drafts. For an example, look at the link I provided earlier to Pasi's AD notes. Possible remedies: more active shepherds in WGs that have this problem, creating a stall state to more accurately reflect where the document is, sending more mail to the WG list to make sure everyone knows what the situation is, and escalation path via Russ in cases ADs are not responding. This is all relatively easy to do.
2) The quality at Publication Request is too low. Do we have this problem? I think we do. Clearly, more work would have to be done at the WG stage, or perhaps some late effort should be moved earlier.
3) Discusses that raise non-issues. There are some of these. Probably one third or so of all Discusses are cleared in the first telechat without any changes, after some misunderstanding or lack of information is clarified. But that's fine. But then there are some cases where a non-issue Discuss survives further and causes an unnecessary change in the document. IMHO a small number, but anyway. The only way to deal with this, however, is the constant discussion between the IESG and the WG (or the greater community) about what issues are significant enough to warrant a mandatory fix. Pushback from the shepherd and the sponsoring AD would be helpful here, obviously only when the pushback really is warranted.
4) Pace of progress through the process is slow. Do we have this problem? Yes we do. Possible remedies include fixes to the specific problems, such as the three ones above, but also setting expectations. Hence the suggested target times. I do realize that we have no means to enforce these in a volunteer organization. And that there are legitimate cases where both ADs and authors may be unable to follow the timelines. However, this does not mean that we should forget setting goals or remind people if they don't reach them.
I realize that #2 is probably the biggie. But lets not disregard the other ones either, particularly the ones that are easy to do...
Jari