Much of the meeting content is better obtained from the slides and the
recordings (including transcripts - you can see the transcripts on youtube).

This minutes only tries to capture the gist of  mic discussions.

Update on the base spec
-----------------------

Jordan will reach out to applicability statement draft authors
on some points to be addressed there.

Tony (Juniper):

Thanks Alvaro for going through this probably one of the largest specs that
you have to suffer through.

Alvaro (AD):

"suffer" is the keyword :-)
Jim (new AD) is taking over tomorrow; I will keep looking at this stuff
little by little until he catches up so that there will be graceful
handover.

Tony:

There are freertr, python (Bruno) and Juniper implementations.
KV registration is where the most action is; Juniper is active in
auto-evpn and auto flood reflection.

Key-value Registry
------------------

Tony:

should have an update on auto flood reflection as well -
good amount of work happening there.

Jeffrey (co-chair):

We did a WGLC before; given the recent changes should we do another
LC after base spec is finally done?

Tony:

No rush ... we get the base spec done first.

Sandy (ZTE):

Better to add some examples to show how to register some values.

Tony:

You can find examples in the auto-evpn spec and applicability statement.
I think there are already kv-registry considerations there.

Charter discussions
-------------------

Alvaro:

If you're going to submit after the current work is done you don't need to
list the work already in the charter.
Of the three current items you listed, the YANG is the last one and you don't
have to wait for that one to be done. When the base spec and applicability
statement are in the IESG you can probably submit the charter.

Jeff (co-chair):

Timewise it's probably between SF and Prague IETFs.

Tony:

The list looks right; PGP are still in the schema so the hooks are there
for it to be easily revived.
Leaf ring works with current ZTP schema but one optimization could be done
as ZTE suggested.

Alvaro:

"Using ZTP for ISIS auto flood reflection provisioning" is more about
"enabling ...". It's ok to work on that in this WG but the charter should
say you're going to talk to LSR and get agreement so that they don't get
surprised.

Tony:

The draft was already presented in LSR. You're right that we need to discuss
this there.

Sandy:

Maybe more informational documents on how to deploy those new features.

Jeffrey:

Will keep in mind of that; the new standards track specs can also cover that.

Jeff:

We'll keep updating as we go; I do expect new work items to come up.
Given the machine learning OpenAI with hundreds of thousands of nodes,
there is not better protocol then RIFT.

Tony:

Sandy has a fair point, with all these new features it's good to have documents
describing how they're deployed.

Chicken-and-egg problem: w/o implementation you don't have much to consider.
Go implement and play with it.

Jeffrey:

We're talking about an interop in Prague hacktahon.

Tony:

But that's only for the base spec. Really encourage more implementations on
additional features.

Jeffrey:

This is just initial proposal - please review and comment.

Jeff:

Alvaro - we really appreciate your insight with regard to the updated charter.

Jeffrey:

Yeah ... Alvaro chartered this WG and this is also my first time being a chair.
Thank you for your guidance and for your thorough review. We know how much
effort you put in and how many issues you found. Really really thank you.

Xuesong (Huawei):

The proposed charter seems quite broad ...  items should be more specific and
from routing protocol point of view.

Jeff:

The initial drafts are already there for each item. We appreciate WG review of them.

...

Again Alvaro - special thanks from the chairs and every participant.

Alvaro:

Thanks everyone for doing the work and I hope we all keep working more on this.