W3C

- DRAFT -

SV_MEETING_TITLE

26 Sep 2018

Attendees

Present
Regrets
Chair
SV_MEETING_CHAIR
Scribe
jfontana

Contents


<plh_> https://lists.w3.org/Archives/Member/chairs/2018JulSep/0139.html

<jeffh> invite zakim?

Tony: philippe is on and will fill us in

PLH: we looked at reuest and we have a few questionsa nd maybe one major concner.
... first is the one that could be major, it is about the extensions.
... understanding we when yo moved to CR if any extensions would be dropped or made informative.
... they were kept as optional

tony: let me address. Giri is pulling together set of arguments for arguments.
... that will be presentedat the TPAC timeframe.
... we will argue all the extensions should be informal and optional
... we beoieve we have basis of argument

plh: hav eot have a phone call wit the director.
... it sould help to organize a phone call on that.
... the other ones are not as critical
... we notice that the spec is using an IETF draft for token binding.


.we did see the email from the chair of that group, but we don't know the status

self-issue: they are RFC , with the editor and expect them in days.

plh: you have said enough, we believe you
... tahnk you
... the next question was, want improvement of working on slight way we talk about what issue gaining or losing focus on document, change is paragraph issue in 5.6 of spec.
... say the issue will be updated.

self-issue: fine for you to do that

tony: it is editorial change.

plh

plh: we address lot of issues that are in edtiiors draft , update those links to the current draft.
... if there are any links that would become broken in the operation we will come back to you , OK?

tony: good to me

plh: build consistencies in your references.
... last point is interop report.
... we received this information. we looked at it and we had some tests in wpt on web auth and most of those are failing. are those tests relevant.

tony: this is the interop. getting it green for Edge vs. the red status

akshay: adam's list.

<plh_> https://wpt.fyi/results/webauthn?label=stable&aligned=true

?

tony: answer is they will not fail. we are updating.
... we will update the matrix for the test that failed

plh: ok. it would be nice to have to updates done when we talk to the director.
... i tried to run all those tests and I was able to get into green, we thought the tests may not have been run properly.

skshay: which link are you talking about, which test

plh: I put the link in IRC
... we need to run these tests.

tony: this may also have to do on how we are running the tests.
... we should have a phone call and pull in Adam.

plh: can you generate a repo to look at the conditions when red, when green
... so this is it.
... we need phone call to talk about the extensions.
... mike jones, sam, giri, co-chairs,
... and maybe jeeH

jeffH: sure.

plh: timeline?

tony: we would like to have this by next week some time.
... i need a few days to talk to giri.

plh: OK

tony: talking next week.

jfontana: monday, tues. wed.
... days we need to do it.many headed to FIDO plenary

tony: and we have issues with the tests. we should have a separate call with Adam
... let me talk to adam

maybe we don't need a call.

scribe: we will tried to get this cleared up next week

plh: yes

<jeffh> https://github.com/w3c/webauthn/pull/1082

emil: JeffH are you suggesting we restructure the algorithm step? or is it a counter issue?
... the server might have already identified the user, or not.

jeffH: it is just the terms that I am concerned about.

emil: not sure how to get around owner / user

jeffH: the term identified, we don't have that notion defined. don't know what account is being used.
... it is RP specific.
... the term identified. did we mean user verification. no.

emil (elundberg): you could have the request refer to values that were used in the request

jeffH: there is not a step to verify the credential source..
... we need steps for the RP on creation.
... one could argue we punt this to level 2

elundberg: it wold be a technical issue

JeffH: i will try to propose text today.

agl: the HMAC secret extension did you use hypen or underscore

akshay: have not seen hypen before, but it looks like it is supported

agl: I think we will support it.

akshay: it is a hypen.

agl: also #1050 I updated it, it is transports in registrations

akshay: i think this looks fine, let me look at it.

rssagent, draft minutes

<jeffh> looking like not....

I think you are right, it looks like Zakim was not started.

suggestions on what to do?

<jcj_moz> copy/paste to the ml, probably

<jeffh> cut'n'paste the above into email msg to the list entitled something like "esatz minutes of webauthn meeting 26-Sep-2018"

<jeffh> ersatz i mean

thanks

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.136 (CVS log)
$Date: 2018/09/26 20:27:14 $