Re: Tricky cross-area topics

[Mark Nottingham <[e-mail]>]

Random thought -- we make WGs jump through a minor formal hoop -- the shepherd writeup -- before going to IETF LC, to make sure they've done everything the need to and save later work.

It might be good to start doing something similar when they *start* a new document, to make sure they take things like this into account. Otherwise, I suspect this wiki page is going to get... dusty.

Cheers,


> On 4 Oct 2019, at 10:28 pm, Suzanne Woolf <[e-mail]> wrote:
> 
> 
> 
>> On Oct 3, 2019, at 3:29 PM, Adam Roach <[e-mail]> wrote:
>> 
>> Working group chairs --
>> 
>> In an attempt to reduce the number of "late surprises" with documents found during IESG review -- that is, showstopping issues with uses of technology that can sometimes require non-trivial reworking of protocol mechanisms -- the IESG has gathered together a list of topics that frequently trip up document authors and occasionally entire working groups.
>> 
>> The hope is that Working Group Chairs can keep an eye on these topics as new work starts up in their working groups, so that appropriate experts can be looped in early in the process, thereby avoiding late surprises during IETF last call and IESG evaluation.
> 
> I like this idea a lot, thanks IESG.
> 
>> 
>> The high-level list lives at <https://trac.ietf.org/trac/iesg/wiki/ExpertTopics>. The intent is that this should be fairly easy to scan to see whether a document under consideration for adoption touches on any of the related technologies. Each area also has its own slightly more detailed page, linked from this list, that goes into the technology areas in a bit more depth; but the intention here is that anyone using a listed technology should reach out to a directorate, expert, or area director in the related area for guidance.
> 
> Please add "special use names" to the bullet on DNS. The classic case is the year-long nightmare around “.home” in RFC 7788.
> 
> I wrote a draft (and then have lately failed to update it from comments, but will before Singapore) suggesting some guidelines for IETF WGs to use in picking special use names, or finding alternatives, if it appears that a protocol “needs” a hard-coded domain name. There are some dragons, particularly if people want “a TLD” for their protocol.
> 
> 
> Suzanne
> 

--
Mark Nottingham   https://www.mnot.net/