On 12/26/2015 03:36 PM, Abdussalam Baryun wrote:
But mostly I think that WGs need to have their work subject to
formal external review much earlier in the process than Last Call,
particularly when the WG's work has the potential to impact other
WGs or other areas.
Easiest way for cross reviews are in meetings
I disagree. Though some amount of cross-area review does happen at
meetings (and if it weren't for meetings, there would be considerably
less), there are several problems with relying on meetings for
cross-area review. The first is that the meetings tend to be scheduled
in area tracks without much regard for the potential for one WG's output
to cause problems in a different area. The second is that the way we
conduct meetings is itself hostile to cross-area review. Though people
from other areas do bring up concerns at meetings, there is a strong
tendency to view those people as "outsiders" who "haven't done their
homework" (perhaps because they don't use the same technical terms as
the WG participants, and/or they just learned of the problem the group
was causing a few minutes before and haven't had the time to read the
group's dozen internet-drafts). And those outsiders are also viewed as
taking up too much precious meeting time which could otherwise be
devoted to more PowerPoint presentations.
(Which is a related problem - IETF working groups are producing too many
documents for the number of participants, resulting in many documents of
low quality and/or low relevance which IESG mostly feels obligated to
publish. And IETF treats that as a sign of success).
Keith