Re: [Sidrops] request for call for Working Group adoption draft-spaghetti-sidrops-rpki-validation-update

[Christopher Morrow <[e-mail]>]

nick got me before done...

On Thu, Apr 22, 2021 at 1:54 PM Nick Hilliard <[e-mail]> wrote:

> Christopher Morrow wrote on 22/04/2021 18:46:
> > Not a lot of follow on this... (well, public follow up)
> >
> > So, I propose this course of action:
> >    1) this document serves as a 'requirement (and possible solution)'
> > that sidrops used to identify a problem
> >    2) send this along as a query/question to the protocol group (IDR)
> > with a request to:
> >       a) say: "go forth, this change seems ok, it's small and such"
> >       b) say: "Good googley-moogley! what are you crazypants people on
> > about?? of COURSE this needs to be reviewed... please have AD-foobar
> > spin up SIDR to properly handle this!! TUT-TUT! How cloud you conceive
> > of a world other than this!!"
> >        and answer back in 2wks time, one fort-night! and two farthings!
> > (or something, farthings are metric)
>
> do we need to involve IDR?  Hard to tell really. Will they get upset if
> they aren't consulted?  (+ will they get ideas if they are?)
>
>
I believe the process we were pitched when SIDR was --'d from existence was:
  "If you (sidrops) see a problem, you should write a document that serves
    as the problem description and requirement for fix-action. Take the
document
    to IDR, who'll promptly do the 3-stooges: "not it" maneuver and hand us
off to
    the AD who will SIDR++ and then we can chat about the problem in the
SIDR
    venue, with possible protocol changes documented/implemented there."

I was proposing that the current document under discussion seems like the:
  "there is a  problem, we have a requirement to fix it, kthxbi!"


> > provided no one screams at me about this WHILE I WRITE THE NEXT EMAIL..
> > we can send an adoption call/etc in 2wks time OR jump back into our SIDR
> > pantsuit and get to the business of business.
>
> sounds good.  The proposal doesn't seem to be horrible, several software
> stacks have implemented it _and_ the rpki hasn't fallen over yet, so it
> seems like so far, it's not a regression and it seems to fix something,
> so ... yay?
>

ok!