Re: [Idr] Martin Duke's No Objection on draft-ietf-idr-ext-opt-param-11: (with COMMENT)

[Martin Duke <[e-mail]>]

Like I said, there are two reasons not to I can think of:
- To test support of this option on hosts across the internet
- To encourage hosts to handle it properly.

On Wed, Apr 21, 2021 at 2:57 PM John Scudder <[e-mail]> wrote:

> Hi Martin,
>
> > On Apr 21, 2021, at 1:39 PM, Martin Duke via Datatracker <
> [e-mail]> wrote:
> ...
> > ----------------------------------------------------------------------
> > COMMENT:
> > ----------------------------------------------------------------------
> >
> > I agree with Rob. Though not at all an expert on BGP, I think that if
> BGP nodes
> > MUST NOT use the new encoding for smaller parameter sets, that is going
> to make
> > it harder to test for the capability on the internet and reduce the
> incentives
> > to support this specification. I would suggest that instead they SHOULD
> use the
> > RFC4271 encoding, possibly discussing the reasons one might not.
>
> As I mentioned in my reply to Rob, I actually can’t think of a good reason
> for them not to, which is why it’s a MUST; as I mentioned in my reply to
> Warren, the reason to consume the new format nonetheless (other than
> Postel’s Law) is to enable potential future deprecation of the old format
> (yes, I’m an optimist).
>
> —John