Re: [core] SID open comments/discussions before WGLC2

[Ivaylo Petrov <[e-mail]>]

Hi Michael,

Thank you for your feedback! During the CORE interim today we briefly
discussed the SID naming issue and what Carsten suggested was to officially
use the name YANG SID (or YSID), but in case there will be no confusion, to
abbreviate this to just SID. That way people looking for YANG SIDs will
always find the relevant documents, but we will not need to do too big of a
rework of the drafts and tools minimizing the risk of mistakes. How does
that sound to you?
Best regards,
Ivaylo


On Fri, Jun 5, 2020 at 2:18 AM Michael Richardson <[e-mail]>
wrote:

>
> {Thanks for the update.  While many people seem to have gotten COVacations,
> it seems that those of us who already worked remotely, just got more work
> :-}
>
> Ivaylo Petrov <[e-mail]> wrote:
>     > My understanding is that with the latest revision -13, we have
> addressed
>     > all pending comments (most notably the early allocation one). The
> only
>     > items that were mentioned and not yet in the draft are:
>     > - SID renaming to YSID suggested here [1]. I have discussed with
> some of
>     > the other authors and we believe that could have an important
> negative
>     > effect on existing code, plugins and changing the name now might
> bring
>     > confusion in the community. We would like to have a better sense of
>     > people's opinion on that.
>
> I think that everyone with a stake in the ground is in this WG, so I don't
> think it's an issue.
>
> I am not super enthusiastic about YSID as a name/TLA, but maybe it will
> grow on me.
> (I try to pronounce it, and I come up with "Why-Sid", which becomes
> "wisest"
> and then reminds me of "Weizmann", as in "Weizmann Institute of Science"
> for
> some reason.)
> BUT, I can live with YSID as a name, and I think it's less confusing then
> SID.
>
>     > - Moving some content type into yang-cbor and sid drafts
>     > Problem statement: if anyone would want to use CBOR + SID
> representation
>     > or CBOR + Names representation of YANG data, they need to have
> normative
>     > reference to the CoMI document, which might appear unexpected.
>     > Proposed solution: as discussed [3] we can either move the media
> type and
>     > content format registrations in the yang-cbor draft, but then it
> will have
>     > a normative reference to the SID draft that we really wanted to
> avoid, or
>     > we register CBOR + names in the yang-cbor and we register CBOR + SID
> in the
>     > SID draft, but then we have normative reference in the opposite
> direction -
>     > SID to yang-cbor. It appeared that people were more in favor of the
> latter
>     > approach, so the changes that I thought this would need could be
> found at
>     > [4]. I am not perfectly happy with this, so I would like to know the
> WG
>     > opinion on how important this problem is to be resolved and whether
> we go
>     > with the first option or the second one.
>
> Yes, I agree that neither situation is ideal, and what you describe in [4]
> seems find.  I agree that a reference to COMI would be surprising.
>
> --
> Michael Richardson <[e-mail]>, Sandelman Software Works
>  -= IPv6 IoT consulting =-
>