Re: [Idr] WG LC on draft-ietf-idr-wide-bgp-communities-06.txt (2/4/2022 to 2/18/2022)

[Robert Raszuk <[e-mail]>]

Indeed Sue - 100 % with your assessment on this.

In fact the current implementation report proves very well Job's point that
atoms can be defined also as separate documents.

The only question which I think Job brings to the table is if currently
defined atoms in the WGLC draft could stay or should be removed.

My personal recommendation is to keep them in the spec. Reason being that
authors went via extensive discussions with community which lead to define
and document those atoms. Keeping them in the document also for one allows
any implementation to catch up while does not close the door (as proven) to
define new atoms.

Kind regards,
Robert.


On Tue, Feb 8, 2022 at 12:41 AM Susan Hares <[e-mail]> wrote:

> Robert:
>
>
>
> I agree that the concept of atoms has clearly been proven by the
> RouteAttr.
>
>
>
> The RouteAttr_  in the implementation report is a hack to work around the
> wiki formatting issues.
>
>
>
> Sue
>
>
>
> *From:* Idr [mailto:[e-mail]] *On Behalf Of *Robert Raszuk
> *Sent:* Monday, February 7, 2022 11:49 AM
> *To:* Job Snijders
> *Cc:* Keyur Patel; Susan Hares; idr@ietf. org
> *Subject:* Re: [Idr] WG LC on draft-ietf-idr-wide-bgp-communities-06.txt
> (2/4/2022 to 2/18/2022)
>
>
>
> Job,
>
>
>
> https://datatracker.ietf.org/doc/html/draft-ietf-idr-rpd-15#page-8
> defines two new atoms and documents its implementation. Does this mean that
> atoms from the original spec are to be removed ? To me this proves that
> core of the spec is extensible which is a feature not a bug.
>
>
>
> Thx a lot,
>
> R.
>
>
>
> On Mon, Feb 7, 2022 at 5:42 PM Job Snijders <[e-mail]> wrote:
>
> Dear Sue, others,
>
> Thank you for your message.
>
> On Mon, Feb 07, 2022 at 11:30:40AM -0500, Susan Hares wrote:
> > Job:
> >
> > <WG Chair>
> > Thank you for letting me know there are broken links
> >  on the implementation page.  I believe I have addressed these issues.
> > Please reload your page and let me know if there are more.
> >
> > I hope you will provide feedback on atoms 1-8
> > as optional  features that are not implemented.
> >
> > In the past, the IDR has made the following choices on
> > 1) remove from specification (knowing may come back as a BIS)
> > 2) hold the draft for implementations of atoms 1-8.
> > 3) keep optional features in the text, but note lack of
> >     implementations  (rare, but possible if WG desires).
> > <WG chair off>
>
>
> As 'Atoms' appear to be a core aspect of Wide Communities, so it seems
> quite awkward to publish this document - without any implementation of
> the concept of Atoms!
>
> To me it seems that 'option 2 - hold the draft for implementation' is
> the simplest path forward, or as secondary choice 'remove from spec'.
>
> Publication of internet-drafts which lack implementations, in my opinion
> sets dangerous precedent for IDR; I hope the WG appreciates my
> insistence on this matter.
>
> Kind regards,
>
> Job
>
>