ietf
[Top] [All Lists]

Re: call for ideas: tail-heavy IETF process

2013-05-03 10:40:45
On Fri, 2013-05-03 at 14:27 +0100, Stephen Farrell wrote:
On 05/03/2013 01:59 PM, Thomas Narten wrote:

If you look at the delays documents encounter (both in WG and in IESG
review), the killer is long times between document revisions. Focus on
understanding the *why* behind that and what steps could be taken to
make improvements.

Good point. I guess the obvious answers are "not enough
cycles" 
and not enough "concentrated cycles" - designing and documenting a
protocol in my experience requires a longish period of undisturbed
concentration by the person with the editing crayon, and responsive,
deep and immediate review by co-authors to generate fast turn round
reviews of versions.
Its hard enough keeping the in-brain implementation of the protocol
running accurately day-by-day.  If there are gaps of weeks rebooting the
implementation wastes precious concentrated time.  Probably putting a
few people with no other commitments in a isolated space generates
fastest results. Maybe we could get ISOC to provide a retreat space (or
one per continent) for protocol designers?
and, for newer authors, uncertainty about how to
get stuff done, but are there other less obvious answers?
(Input here might really help the IESG discussion btw since
in the nature of things, we're less likely to realise what
newer or less frequent participants find problematic.)
One thing that might help:
We have directorates and review panels.  Can we distil any/more of their
wisdom into checklists/guidance for protocol writers?
Some is already there (not a complete list):
BCP 72 for security considerations 
BCP 41 on congestion control
BCP 61 on Strong Security Requirements
Some stuff in RFC 6385 for gen-art

[Some of this is getting a bit long in the tooth.]

So what do you ...
- think about when designing the security aspects/transport congestion
aspects/... of a new protocol?
- what triggers alarm bells/gold stars when you are reviewing the
general principles/security aspects/congestion/ABNF/state
machines/mib/xml/extensibility/.... of a draft?

Regards,
Elwyn


S.