Re: [regext] list of documents to consider for working group adoption

["Patrick Mevzek" <[e-mail]>]

Hello,

Sorry to be probably again in the minority and the dissonant voice here, but I do not understand this:

On Fri, Dec 7, 2018, at 10:49, James Galvin wrote:
> The purpose of the list was to include everything that exists or that 
> has passed through the working group.  I have to admit it was more than 
> I expected when we created the list.
> 
> In any case, the right thing to do is simply not vote for any document 
> you do not want to move forward.

To me, this process feel completely backwards :-(

As described, it makes sense to me if:
- we have a wide community of people
- with far enough time on their hands
- to read ~20 technical documents
- compare their merits
- and finally select 5 to "promote" to push to the working group
- of course all in some specific tight deadline.

I believe we are absolutely in the opposite case and I fear the outcome of this process will not have the expected results.

First, I think the burden should be on the authors to at least present the document on the list (which is not even the case for some of them), in order to convince people that there is indeed a problem to solve (first important point) and that their solution or beginning of a solution makes sense. If their arguments are clear and correct, people would then be enthousiastic to add them as working group elements and work on them.

Instead by just trying to be exhaustive and let people choose among everything you achieve two things:
- you show each document as similar to others in term of usefulness that is maturity, scope of the problem, precise solution to it, etc. (and they are obviously not all equal on these points)
- you accept that documents are basically worked on elsewhere, never discussed here, and coming late (which means it may be more difficult to work, as a group, on them because many design choices would already be set in stone by then). Which has to me the very unfortunate result that the working group just becomes an IETF stamp for an EPP extension.
Even if the same people as here are working on the document elsewhere before presenting it here, what is then the value or added value of this working group, if it is basically just to make sure it fits an RFC structure, follows all guidelines, register what needs to be registered at IANA, etc. which is basically no more an engineering task but merely an editorial one?
This is not the only reason but I think this also explains in the very low participation we see and the difficulty to find editors, shepherds, etc. (and of course it is a vicious circle as the low participation here could be taken by some as an indication that they should work on their documents elsewhere to make progress).

Second, beside authors of course which are already motivated to do so, anyone may be free to promote or vote on any document of couse to be a working group document, but what would that mean (for the working group and the specific documents)?

Instead, why not do like it is done in other working groups and ask people to speak for a document **if and only if** they kind of informally pledge to work on it in some way which may mean:
- reading it, and the future other versions
- being at least able to assess if the document is good enough to go to WGLC for example
and/or to summarize it broadly (problem to be solved, strong points of the solution offered, points remaining to discuss, other options)
- provide feedback, if possible, at various levels (either just alternate ideas or specific detailed implementation point to work on, etc.)
- act as an editor (non-technical feedback)
- pledge or express strong interest into implementating it or putting enough energy to convince a third party to implement it
- identify that it may be related to some other kind of document (in this working group or not), or that other actors may already have solved the same problem or working on it, etc. and trying to liaise between everyone
- etc. (there are many ways to participate, sometimes just sending emails to the authors to ask where they are at, what is blocking, etc. can give the authors a boost of motivation - seeing that others care too about the document - to work further on it)

This would also clearly show interest for it: if people are saying "yes, I agree to spend some of my time making sure this documents advance as an IETF document" then it proves real usefulness and support by the "community" for it. A very useful information to be added later on in the shepherd write-up ;-)


Also the discussion started already previously (but did not go very far I think) and I think we (I) suggested already that having an Implementation Section or at least clear intentions to work on implementations or asking others to work on some could be used, among other factors, as a basis to see if a document should be considered by the working group or not.

To be honest right now I prefer to devote the little personal time I have to try implementing some of the documents that I find interesting (as implementing them raise all sorts of feedback) instead of reading completely external documents and then compare all of them together while not be sure to even look at some of them again in the future. Which is why I won't vote for any document right now because some are not passing the preconditions I listed and for others I can not guarantee working on them at any level so I believe my vote to be kind of worthless for the working group.

But that is just me so if the process above works for everyone as is, do not loose time trying to convince me, it is not very important, and I can always comment later on the documents I would have time to work on.


-- 
  Patrick Mevzek