Minutes interim-2018-alto-01: Tue 06:00

Meeting Minutes Application-Layer Traffic Optimization (alto) WG
Title Minutes interim-2018-alto-01: Tue 06:00
State Active
Other versions plain text
Last updated 2019-01-02

Meeting Minutes

ALTO Virtual Interim Meeting
Dec-11-2018 1300 - 1500 CET
Attendees: ~16

Raw notes taken by Kai Gao and Vijay K. Gurbani

Intro and agenda bash (slides at [1])
 - alto-cost-calendar and alto-xdom in IESG evaluation.
 - alto-incr-updates in WGLC
 - as WG docs: alto-cdn-fci, alto-path-vector, alto-unified-props-new

 Agenda bashed and accepted.  Plea by chairs to focus on finishing current
 charter work as soon as possible.

alto-cost-calendar (slides at [2])
 Sabine briefed the WG on the status of the draft.  It appears that IESG
 needs a revised I-D to take care of two DISCUSS items: one is related
 to IPR and the second is related to design issue (see slides 4-6 of [2]).

 Some discussion ensued on how to proceed with respect to soliciting the
 changes resulting from the DISCUSSes in the WG.  It was decided that a new
 I-D will be produced and a new WGLC will occur in the WG to ensure that
 the changes from the existing DISCUSS are okay.  Subsequent to the new
 WGLC, the document will be sent back to the IESG.

alto-unified-props-new (slides at [3])
 Jensen presented the work.  Sabine had some comments on the difference
 between identifiers and property names.  Discussion ensued at mic, and
 it was subsequently decided that:
 - Sabine to post the question on mailing list;
 - Sabine and Jensen drive consensus to close/answer the question;
 - A new I-D reflecting the consensus appears in the archives;
 - Pursuant to above, chairs to issue a WGLC on the new I-D.

alto-incr-update-sse (slides at [4])
 The draft has been in WGLC for a while.  Richard noted  a couple of places
 where the WG attention is needed:
 - Whether the new Section 11 should be moved to the appendix?
 - What to do if the media type has commas (which are not allowed in current
 Regarding the first issue, some discussion ensued, but no particular
 consensus reached on whether Section 11 should be moved.  Richard to
 close this issue on the WG mailing list.
 Regarding the second issue, Richard to send email to IANA media type expert
 and reach an equitable solution.

 Pursuant to the above two issues being closed, the WG can move this draft
 out of WGLC and towards IESG.

alto-path-vector (slides at [5])
 Richard went through the work.  Some discussion ensued between Richard and
 Kai on why costmap/propmap in the request if the response is always
 multipart?  In the interest of time, this discussion was requested to be
 moved to the list.

 Vijay noted that the draft needed an in-depth review before we move to
 WGLC.  It was pointed out that Sabine has done such a review.  Will move
 to WGLC once the discussion between Richard and Kai comes to a consensus.

alto-cdni-request-routing-alto (slides at [6])
 Shawn presented the document and informed the WG that he may not be
 able to finish the work as the draft may benefit from some new
 optimization work.  There was general discussion that coalesced around
 the fact that Shawn should wrap up this document in its current form
 to attend to cdni footprints and leave optimizations for a later revision
 to the RFC-to-be.

 That seemed to be agreeable to Shawn.  Shawn to drive the work to

The remaining drafts were individual drafts.

zhang-alto-multipart (slides at [7])
 Jensen presented the work and ask two questions:
 - Should we design a domain-specific query language for ALTO?
 - How interested would the WG be in such work?

 Clearly, what to do with this work is a question that can start to be
 answered once ALTO is done with its current charter.  The work looks
 interesting in general.

xiang-alto-multidomain-analytics (slides at [8])
 Qiao presented the work.  Vijay commented that this is really interesting
 work that has informed both the academic community (SC 18) and the

 Mirja noted that the work is not in the current ALTO charter, but that
 due to the applicability of the work to ALTO, the work could conceivably
 proceed in the IESG stream.

lachosrothenberg-alto-brokermdo (slides at [9])
 Danny presented the work.  Vijay noted that there are some concrete
 extensions to ALTO worth exploration.  However, any such discussion will
 need to be undertaken after the WG is done with the current charter