Re: Quality of Directorate reviews

[Michael Richardson <[e-mail]>]

Carsten Bormann <[e-mail]> wrote:
    > But please do keep in mind that after having worked a couple of years
    > on a document, the WG might no longer see things that a fresh look will
    > uncover.  I think getting that fresh look is one of the great features
    > of the multi-step review process we have.  Also, there is the simple
    > fact of serendipity!  (Which is not very structured, and someone
    > schooled in quality processes will hate, but creating quality is not
    > really a controlled process anyway…)

1) We need these things to happen significantly before WGLC, and the WG has
   to not regard these things as hostile; this means that the review has to help.
   The reviewer has to then become involved in the changes that result.
   (This argues for the "stable document" statement, but I'd way prefer that
   we published stable-document as PS, using some lighter weight process,
   even if it winds up in a new series)

2) The AD for that directorate has to be involved such that the relevant AD
   does not just repeat the entire conversation.

3) We need the time from beginning the document process to the end of this to
   occur a bit faster as authors of a document simply get exhausted.  That
   probably argues for shorter documents.  It also argues for more hands to
   be involved, which might mean to me, lifting the 5 author limit.

--
Michael Richardson <[e-mail]>, Sandelman Software Works
 -= IPv6 IoT consulting =-