Hi David,
I was wondering whether there are plans to actually make a selection out of the different proposals, or see if a design team can be formed that will investigate a way forward that is based on several of theproposals.Quite honestly, I am a bit concerned that continuing work in parallel on all the different proposals might easily create a situation where it becomes very difficult to decide on a single solution as people have invested already quite a bit of their time and energy on a particular track.
My concern tends to run the other way. I think a design team would be premature. As one author, I'm not wedded to my design or to LISP, for that matter. I do think several key things are needed before we get into design committees:
* A clear understanding of which problems each solution solves
* Some experimentation to see how each solution behaves
* Some analysis that looks at stress points with each mechanism
* and a summary of all of this that compares approaches. Robin
Whittle made an attempt at such a thing, but it would be better
when we have all had a bit more hands on.
Now don't get me wrong - I understand that things can work in the small
and not in the large. I think that's why the scaling analysis is
crucial. It needs to take into account motivations and behaviors, which
are always tricky at best. For instance, what if one of these
approaches actually encourages disaggregation into the existing BGP
core? Could that happen?
For my part, I've set out to do a bit of the stress point analysis in the next rev of the NERD draft, but it is woefully incomplete. Also, my goal for that draft is for it to be published as an EXPERIMENTAL RFC. At such time as we actually want to standardize, then I think a design team would be appropriate. I'm just not there yet.
Regards, Eliot -- to unsubscribe send a message to rrg-request@psg.com with the word 'unsubscribe' in a single line as the message text body. archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg