General comment on draft-ietf-problem-statement-00.txt

[Pete Resnick]

On 3/3/03 at 5:16 PM -0800, Randy Bush wrote:

>To: Eric Rescorla <ekr at rtfm.com>
>Cc: Margaret Wasserman <mrw at windriver.com>
>Cc: problem-statement at alvestrand.no
>Cc: Keith Moore <moore at cs.utk.edu>
>Cc: dbh at enterasys.com

[Side note: Multiple Cc lines are a no-no; they screw up some mail 
clients. They're currently on the "accept but do not generate" list. 
I don't know if that was Randy's mailer or the list manager, but 
please look into it.]

>so how do we get wgs to produce production-quality output and get 
>production-quality ietf last call review?

OK, so you're asking for solutions here. Without going into detail 
(since we're only supposed to be talking problems and not solutions), 
I'll offer some short suggestions of directions that could be taken:

1. Tell WG chairs that documents from their WG will not be reviewed 
unless they have done a full technical writeup of the documents in 
question, something akin to what ADs do now.

2. Require WG chairs to solicit document commentary from people 
outside of their area and to include that commentary with the 
submission of the documents to the IESG.

3. For every WG, have an IESG approved volunteer advisor for each 
area outside of the WGs area. The advisor would be responsible for 
keeping up with the documents of the WG and giving input as needed. 
In Ted's parlance, there would be "external stuckees".

To put this back into problem instead of solution talk: To accomplish 
some of these, we would need to address "chair competency" problems 
that the document outlines. Some of the above would need more 
organizational structure, delegation of authority, and other 
structural changes that address some of the other problems that the 
document outlines. And if implemented correctly, some of these things 
might (perish the thought) make it completely unnecessary for the 
IESG to do the kind of intensive detailed review that it is currently 
doing because the problems of quality would be addressed at the 
beginning and middle of the process instead of the end. And perhaps 
(again, perish the thought) that would make it reasonable to "dilute 
the IESGs authority", because solving the problems otherwise would 
make the authority of the IESG unnecessary.
-- 
Pete Resnick <mailto:presnick at qualcomm.com>
QUALCOMM Incorporated - Direct phone: (858)651-4478, Fax: (858)651-1102