-------- Original Message --------
I read Dave's words "clear statement of what actions must be taken to clear the Discuss" not as requiring the specification of a complete fix, but rather as an indication of what needs to happen to the draft. Implementation details of meeting those requirements are left to the WG.
I don't know what Dave meant by his words, but I can say that similar words were being used when I was on IESG, and at that time they was taken to mean a bit more than you assume Dave meant. It wasn't enough to say what was wrong with a draft, the AD was expected to tell the WG what to do. The expectation was often that the AD should supply specific text for minor changes to a document that would allow the AD to clear his discuss. This further implied that any problems that could not be fixed with small textual changes had to remain in the draft - the AD wasn't able to say, e.g. "the entire approach has serious and irreparable technical flaws", or "the protocol appears to be sound but the document is so poorly written that small changes will not save it" - even though one or the other were occasionally the case. And this further led to ADs insisting on small textual changes that had no real effect on the document - mostly to make the ADs feel better for having to approve such poor documents.
Of course, such flaws should have been caught a long time before Last Call. That's one of the big problems with our process - we don't have good mechanisms for catching those problems, particularly when they cross area boundaries. We leave it to the supervising AD to implement any ad hoc mechanisms they choose, but the WGs are able to push back on those because there's no community consensus behind them.
Keith _______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf