Re: Late review management (Re: [Icar] independence of reviews; variability)

On Wed, Mar 10, 2004 04:42:28PM -0500, Margaret Wasserman allegedly wrote:
> While there is no decision making process (hierarchical, democratic, 
> I just get to decide...) that achieves good results all of the time, 
> the consensus-driven process has worked well for the IETF over the 
> years.  By failing to trust it, we don't actually move to another 
> effective decision making process, we just break the one that we have.

This whole exercise was triggered because enough people felt that
frequency of mistakes in the process had reached a threshold and needed
to be dealt with.  As the Internet has become more important, the level
of mistakes that it tolerable has shrunk to close to zero.  Also the
level of experience of the participants has decreased.  I agree in
principle, that we want to keep our basic processes, and I trust the
basic process in principle, but we must make some changes to ensure
quality.  We want to add to it, e.g. a few review processes, preferably
in parallel with other work so they don't introduce delay.

> I think that this same line of thinking applies to review...  We 
> should put into place the mechanisms, tools, training, etc. to 
> improve the community's capacity to provide quality review.  Perhaps 
> we should provide mechanisms that help WGs' find and recruit 
> reviewers.  We could even develop some guidelines about what type and 
> quantity of review makes sense at each level.  But, ultimately, I 
> think that we should trust that our WGs (and our WG chairs, document 
> editors, etc.) actually _want_ to  produce good quality, 
> well-reviewed work.  We can improve their ability to do that by 
> giving them better tools, but we won't achieve anything by trying to 
> enforce quality through "better" rules.

I don't doubt that WGs want to produce good work.  It's not a question
of motivation, but of capability (WGs and ADs both, and toss in the IAB
members as well).  We need to bring more minds to bear on decisions --
that is, review them more -- because the stakes are higher and our
ability is not.  Conceptually we are not giving the WGs better tools, we
are giving the IETF better tools.

swb

_______________________________________________
Icar mailing list
Icar@ietf.org
https://www1.ietf.org/mailman/listinfo/icar