Re: [I18nrp] Additional input needed for i18nRP BOF

[Adam Roach <[e-mail]>]

On 6/6/18 4:19 PM, Peter Saint-Andre wrote:
> On 6/6/18 3:06 PM, Adam Roach wrote:
>> I trust that you will attempt to
>> provide these additional details?
> Adam, what are the IESG's requirements and expectations regarding
> creation of a new directorate? Is it expected, for instance, that a
> charter of some kind will need to be provided and approved? Guidance
> would be helpful.

With the caveat that I may have overlooked some bit of formal process...

In the general case, the creation of a new directorate simply requires 
convincing an area director that such directorate would serve a useful 
purpose. While directorates don't have charters per se, as a practical 
matter, they need to have a reasonably complete description of their 
purpose and role.

I'll note that directorates have a broad range of formality and document 
coverage associated with them: on one end, SECDIR and GENART have active 
secretaries who assign every document during IETF last call to a 
volunteer from a large pool of specialists (SECDIR) or generalists 
(GENART). On the other end, the SDPDIR has a relatively small pool of 
self-directed experts that authors, chairs and area directors can reach 
out to for advice and review. The SDPDIR does not proactively review 
documents.

In the specific case of establishing an i18n review directorate: based 
on the comments made so far, I think we would want a carefully defined 
process (possibly documented in an RFC, but perhaps just part of the 
directorate description) that described what role the directorate would 
play in the development and publication of specifications.

We would also need a plausible group of volunteers to execute that 
process. For example, if the proposed process were to review every 
document, in the style of SECDIR and GENART, I'd want to see a group of 
i18n expert volunteers (or volunteers who are willing and able to 
educate themselves into being experts) of roughly the same size as those 
directorates. For reference, the SECDIR has 64 members, and GENART has 
27. By contrast, if the role were to provide feedback only as requested, 
it could be similar in size to the SDPDIR, which has seven members.

In practice, I suspect the process the interested parties might 
reasonably settle upon would be somewhere in between, with someone 
(possibly a rotating duty) providing light triage on documents during 
IETF last call for the purposes of identifying which have an 
intersection with i18n, and a pool of dedicated reviewers to whom 
documents that do are assigned. The number of people such activity would 
require to succeed is, of course, a judgement call, but based on the 
size and function of existing directorates, I think we can safely say 
that it is significantly more than seven, and somewhat less than 27.

/a