Hi Dave,
Let me try the simplest summary possible:
If someone has the authority to block the long-term work of a group of IETF participants, they have an *obligation* to take their concerns directly to those participants and engage in a direct process to resolve it.
Authority always comes with responsibility. In this case it should simply be that the authority to block a group has a responsibility to interact with that group.
Directly.
This seems pretty reasonable to me, and I think that this is one of the reasons why the IESG (pre-me) chose to implement the I-D tracker. I agree that it might be even better to generate mail to the WG mailing list (some folks may read the list but not follow documents in the tracker), but I'm not sure about the best way to implement this...
The most common place where I personally block and/or delay a WG document is AD Review. I only issue discusses on a small fraction of the documents that come to the IESG, but I return AD review comments (blocking or non-blocking) on a much higher fraction of the documents that I am asked to shepherd.
For some time, I have been sending my AD review comments to the WG mailing lists, rather than just to the authors and WG chairs, and it seems to work quite well. I also copy those comments into the tracker, so that folks who are interested in the status of the document can find them.
Most of the time, my AD review comments don't spark any debate, but I think that doing this allows the WG to argue with me if they disagree with my comments. It also helps to improve visibility into how/why the document is being modified after the WG has declared it "done". I think that this practice may also increase WG awareness of the fact that there is now an action item for their editor, and that public scrutiny may result in the editor turning the document around more quickly (I have no statistics to back this up, it is just an impression).
I think that it would also be helpful to send IESG discusses and comments directly to the WGs (for all of the same reasons), and there are two ways that I think we could accomplish this:
(1) When an AD has an open discuss that he or she does not clear during the telechat, he or she could send a copy of that discuss to the WG mailing list directly. This is quite direct, but might be a bit tricky in practice due to spam filters, etc. because the AD may not be a member of the mailing list in question. It is also more likely to suffer from human error or omission, because the tracker is set-up to show us the documents for which we are responsible AD, not those for which we currently hold discusses.
- or -
(2) After the document appears on a telechat, the responsible AD could send any remaining IESG discusses and comments to the WG mailing list (probably in a single message), cc:ing the ADs who entered those discusses or comments. I sometimes do this already, and it typically seems to work well. If folks think that it would be useful (for efficiency or visibility), I'd be happy to start doing it for every one of my documents that still has discusses open after the telechat. This approach would also resolve an outstanding issue with the PROTO process, by making it clear when it is time for the WG chairs to start working to resolve the discuss issues. This approach is somewhat less direct, but perhaps more practical than option (1). It may be subject to some human error, but the AD will see the document in an "IESG Review" stage each time he or she enters the tracker, so he or she will have an opportunity to notice his or her omission.
- or -
(3) We could modify the I-D Tracker so that it will send mail to the mailing list at one of two points: (a) whenever anyone enters a discuss position, or (b) whenever a discuss position remains after the documents has been on a telechat. This is both indirect and impersonal, but not subject to human error.
Thoughts? Do other people think that it would help (efficiency or visibility) for all discusses to be sent to the WG mailing lists? Any thoughts on which of the three approaches above would work better?
Margaret
_______________________________________________ Ietf@xxxxxxxx https://www1.ietf.org/mailman/listinfo/ietf