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

[Dave Crocker <[e-mail]>]

Eric,


Avri>> My hypothesis  is that early  reviews, especially if they  are tightly
Avri>> linked with the with the IESG and AD, will be more likely to point out
Avri>> how to come in line with the AD's agenda. 
ER> Excellent point.  I  had assumed that the "earlier,  wider reviews" would be
ER> harder for the ADs to control very closely (for reasons of scale, if nothing
ER> else), but perhaps that assumption is wrong.

Early reviews wind up requiring _everyone_ to put their expectations and
concerns out on the table sooner.

Unless I missed something over the last 15 years, there is nothing that
gives ADs a unique insight into the creation of perfect protocols. They
are typically senior contributors, and that usually does give them
excellent insight, but not unique.  (For these discussions, I cannot
stress strongly enough just how important that distinction is.)

So, the idea that the rest of us all must always march to the desires of
particular individuals -- no matter their title -- goes exactly against
the grain of rough consensus.

By getting everyone's expectations and concerns out on the table early,
disparities can be reasonably negotiated openly and carefully.


ER>   If the effect of early reviews
ER> is help  special interest  groups have  more impact on  the WGs  rather than
ER> less, we won't have made much progress.

One of the areas in which the IETF has demonstrated a consistent lack of
skill is political/psychological gaming. We do best when we do what is
straightforward and constructive. Special Interests that involve
careful, intelligent folk, will almost always be better at the gaming
than the the IETF community. We hold sway over them by doing Good Work.

So let's not be too distracted by such concerns.  Early reviews are
pretty obviously a good thing, for all the reasons folks have been
saying.


ER> This  all  comes  back to  the  underlying  issue  of whether  "the  general
ER> interest"  is best served  by applying  a variety  of perspectives  (which I
ER> think means  a variety of people  with independent perspectives,  and by the
ER> marketplace), or whether it is best  served by having all the decisions made
ER> by a small  group of know-it-alls.  This  is one of the issues  on which the
ER> problem-statement group failed to agree.

The essence of rough consensus is the belief that paying attention to a
variety of perspectives ensures that there will, later, be both the
ability to use the output to satisfy real-world needs, and the general
support for pursuing it.

The problem with giving an individual or small group excessive control
is the likelihood that they will force an outcome that is not, later,
embraced by the folks who are needed to make the work successful (ie,
useful.)


d/
--
 Dave Crocker <dcrocker-at-brandenburg-dot-com>
 Brandenburg InternetWorking <www.brandenburg.com>
 Sunnyvale, CA  USA <tel:[phone]>


_______________________________________________
Icar mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/icar