W3C

– DRAFT –
Web Authentication WG

14 July 2021

Attendees

Present
addison, jeffh, jfontana_, matthewmiller, nsteele, selfissued, wseltzer
Regrets
-
Chair
Fontana, Nadalin
Scribe
jfontana_

Meeting minutes

thank you, Wendy

Charter: what is the target date for approval?

https://github.com/w3c/webauthn/pull/1630

tony: waiting for Mike to get on the call. Looks like he stated his concern in writing.

https://github.com/w3c/webauthn/pull/1622

jeff: I will put a flag on this, I said my peace

https://github.com/w3c/webauthn/pull/1621

elundberg: t his is a work in progresss

tony: shane, are your issues answered?

shane: I will check it again.

https://github.com/w3c/webauthn/pull/1676

tony: work in progress?

jeff: yes, in progress

https://github.com/w3c/webauthn/pull/1425

tony: work in progress

?

elundberg: yes, in progress

https://github.com/w3c/webauthn/pull/1649

elundberg: This would merge into PR #1621.
… wnat to do as part of #1621

jeff: yes, we should do this.

selfissue: I will look at it

jbradley: I will look

https://github.com/w3c/webauthn/issues/1646

agl: think we should wait to end, things will change

Addison: happy to discuss

tony: what do you want us to do?

addision: work out right guidance for language metadata
… some editorial problems

agl: no one is going to implement it

addision: lets work out best way to fix the metadata issue

agl: do PR for the old chain.
… there is one field and can only be one.
… we were trying to accommodate everything you specified.

addison: what is the best way to start

agl: we will adopt what you recommend
… I could trun something into a PR

addison: Ok
… I want to make sure we don't recommend something is a non-starter

agl: it is unlikely for chrome to implement this.
… we have not heard concerns
… I only speak for chrome

jbradley: some issues. RP has to know what serialization

addison: if we come up with proposed it looks like it needs to be a single field

jbradley: we cold have a metadata field

tony: Addision can you open a new PR? Do you want this in a first working draft

tony: we will deal with the five issues after that

<selfissued> I requested changes to the tokenBinding PR in https://github.com/w3c/webauthn/pull/1630#pullrequestreview-706636912

addsion: let us know if you have questions or issues.
… we will track

https://github.com/w3c/webauthn/issues/1640

akshay: we want to do this at scale.
… #1614 is very much separated into its own issue, because it talks about syncing keys
… this is more about how you do the recovery
… #1640 more about RPs

tony: we haven't had this issue like web payments have, but we have had lots of discussion with Anders
… leave this open

https://github.com/w3c/webauthn/issues/1639

agl: the answer is in #1637
… lets wait for comments then we can resolve

https://github.com/w3c/webauthn/issues/1638

tony: this is Anders

agl: #1637 may resolve this
… help resolve this

https://github.com/w3c/webauthn/issues/1637

agl: this is quite large, people need time to read
… goal is to use web authn to replace passwords
… we think backup in on phones. we layout design of how phones can be used an authenticators
… there are ecosystems. apple, android, windows, etc.

tony: this is across ecosystems

agl: you could send assertion across eco-systems
… see useful changes in web authn
… largest one is conditional UI
… we are updating it.
… adding auto complete
… we think this makes sense
… bits to fill in - want transport in assertion
… what we are saying we will tell websites to put cred IDs in exclude list.
… we will tell website valid state error
… re-authentication is best prractice.
… upgrade to web-authn is a suggestion; read
… akshay has a follow-up and people should read that.

MMiller: lot of these things are addressed, there are some mentions of specific options; are we getting in realm of magical options
… do these options work, that is a question
… report signaling, intent is they are immediate signals to help sync authnticators

agl: I think these options make sense, I don' t think they are magical

agl: whole issues assume discoverable credentials, think this is best for consumers

<jeffh> https://github.com/w3c/webauthn/issues/1637

akshay: has soem concerns, discoverable vs. non-discover; preventing intentional overwrites, but most of signals here we support
… we do have some differences of opinions
… and for some definitions
… by and large we looked at issues, and we noted some patterns we need to solve
… it is mostly about mass adoption.

tony: time frame, seems like big change

agl: we will work on getting everything discussed, some will be longer than others.
… want this to land this calendar year

elundberg: this seems like a reasonable ides, we have to look at it.

jbradley: we should talk about what durable means.

tony: when do we run out of re-charter time

wendy: Dec. 31

tony: what is target date?
… want end of august

akshay: we want to start sooner, sooner the better

tony: try to close on charter at the next meeting.

*minutes updated

Minutes manually created (not a transcript), formatted by scribe.perl version 136 (Thu May 27 13:50:24 2021 UTC).

Diagnostics

No scribenick or scribe found. Guessed: jfontana_

Maybe present: addision, addsion, agl, akshay, Charter, elundberg, jbradley, jeff, MMiller, selfissue, shane, tony, wendy