[rfc-i] RFC Editor structure

Paul Hoffman paul.hoffman at vpnc.org
Thu Sep 25 11:15:10 PDT 2008


At 10:56 AM -0700 9/25/08, Bob Braden wrote:
>Abstractly, what has happened is this: the IAB and IAOC wished to break the
>unitary RFC Editor function into modular pieces, for maximum flexibility in
>contracting them out.  As every computer scientist knows, successful
>modularization requires breaking at boundaries across which communication
>is minimal and well defined.  Breaking off the Independent Submission
>Editor function seemed like a slam dunk for modularization.  But, to my
>surprise, at least, we come to realize that the Editorial Board, regardless
>of whether it is formally or (as currently) informally established, breaks
>the modularity! Your solution was to make two separate Editorial Boards,
>one for each side of the boundary.  (You would presumably give them
>different names). Others (like me) thought this was getting baroque, but
>maybe it is necessary.

Having two boards allows each board to be made up of people with 
different specialties. The folks reviewing independent submissions 
can be technical experts across a wide variety of fields (as they are 
now), but have no publishing experience outside the IETF. The folks 
helping the RFC Editor publish would hopefully have publishing 
experience outside the IETF, but might not be technical experts.

>You seem to be concluding that we CANNOT split off the indsub review
>function, that it is inextricably bound up with the RFC Editor
>function.

It seemed clear from Olaf's message that *he* wasn't concluding that; 
instead, he was summarizing others opinions, stating that they seemed 
not to be the group consensus, and asking if they were indeed the 
consensus.

>I am much less convinced about the modularity of production vs. publication
>functions.

Could you explain that more? It seems that the only binding between 
those two is the production people saying to the publication people 
"here is the exact set of bits that is now officially RFC 7654" and 
the publication people putting them on the web site. That should be a 
trivial communications task. (The publication people should, of 
course, validate that the published bits are the ones they sent.)

--Paul Hoffman, Director
--VPN Consortium


More information about the rfc-interest mailing list