Web Cryptography Working Group Teleconference

28 Oct 2013

See also: IRC log


Michael_Hutchinson, +1.617.253.aaaa, Virginie_Galindo, Wendy, israelh, Sangrae, markw, jimsch, jyates, arunranga, +1.650.275.aabb, rsleevi, +1.703.284.aacc, [Microsoft], rbarnes, vgb, nvdbleek, hhalpin
Jim Schaad


<trackbot> Date: 28 October 2013

zakrim, [ipcaller] is jimsch

<jyates> 617.253.aaaa is jyates

<wseltzer> zakim [IPcaller.a is Sangrae

<wseltzer> scribenick: jimsch

WebCrypto API

<rsleevi> Scribe: Jim Schaad

Will skip the topic on Dan comments since Dan is not present.

Isreal: Questions on text for promises model
... Questions on stylistic for patterns to follow
... Q1 - Using a mixture of exception throughing and promise failures. Seems better to always call reject resolver
... Easier programing since not try catching. Now have unified handling through promises

<Zakim> rsleevi, you wanted to respond to Q1 - exceptions versus failure

rsleevi: Can carry on more discussion on the maiing list.
... One issue with web IDL invlaalid objects are defined as exceptions, so some external things can though exceptions
... Exceptions during the parameter validation, failures on the resolver for the implemention related failures

Isrealh: Q2 - Reconciing the promises model and a stream based model. These are currently orthognal at this time.
... Verify that things may need to change to deal with streams in the future

rsleevi: Streams are still messy and not finalized. Depends on where the stream spec goes

Isreal: Verify that we may need to modify the API when streams become supported. Should potentially note this in the current spec if not already done

rsleevi: Next thing to be worked on. But yes there will be some API changes to make streams first class objects

<rsleevi> ACTION: rsleevi to add note re: streams [recorded in http://www.w3.org/2013/10/28-crypto-minutes.html#action01]

virginia When will the stream sepc be stable?

<trackbot> Created ACTION-120 - Add note re: streams [on Ryan Sleevi - due 2013-11-04].

Isreal: Mid next year maybe
... Q3: In the event model, did we have ability to reuse crypto opertions? Was this true?

rsleevi: No state transition to allow to go from filnish to re-init.

<hhalpin> Quick point re streams - 7.4.3

<hhalpin> "Feature at Risk": http://www.w3.org/2005/10/Process-20051014/tr

rsleevi: Need clarifition regarding parameters for nested streaming.

<hhalpin> We are supposed to exit CR June next year

<hhalpin> so that's cutting it a bit close

<hhalpin> but adding a note re streams as a feature at risk should be fine, otherwise next version.

Isreal: Need to clarify when things go in on promisses and when things need to be broken appart re parameters

agenda, next

PING review

<wseltzer> http://www.w3.org/2012/webcrypto/track/actions/119

wseltzer: PING group has prepared comments on the API. With suggestion that we review their comments - may lead to new privacy considertions
... Most of the issues are in key discovery and pre-previsioned keys
... Need someone to take an action to review and respond to PING

Virginia: Mark should look at the key discrovery items

<wseltzer> ACTION markw to review PING privacy review

<trackbot> Created ACTION-121 - Review ping privacy review [on Mark Watson - due 2013-11-04].

Virginia: Should be the document editors looking at these items

<hhalpin> Its fine to go to Last Call as long as Working Group has closed open issues

Virginia: Target to answer before end of Nov, after F2F meeting
... Should have agenda item at F2F to review privacy aspects

<wseltzer> action virginie to add horizontal reviews, including privacy, to TPAC f2f agenda

<trackbot> Created ACTION-122 - Add horizontal reviews, including privacy, to tpac f2f agenda [on Virginie GALINDO - due 2013-11-04].

<hhalpin> I think to satisfy PING is pretty simple - its mostly moving text from WebCrypto API to Key Discovery

<hhalpin> with whatever otpions are necessary

Technical proposals on Key Discovery API

MichaelH: Current status is ryan and I have been having discussions on techincal merits
... Certificate references is one issue.
... Possible create new interface for certifictes, did not intend to go that far initially
... Looking for stratigies to identify ceritifctes associated with keys held behind the U

rsleevi: Working group not ready to take on the item based on lack of comments coming from the list

<arunranga> +1 rsleevi.

rsleevi: Understand where the use case is coming from, but need to understand more than the one case

MichealH: Some lack of interest on the list for all things

<hhalpin> Should we just keep certificates for WebCrypto v.Next?

virginie: Need to test the interest of the group more formally?
... Look at this at the F2F for assement of interest

<hhalpin> I also might add having less discussion on the list and stopping new features from being added is generally means the spec is mature enough for Last Call.

mark: Looking for concusus on the use cse before studying the proposal in detail

arunranga: : Will ther ebe a dial in number?

virginie: No response to inital request on this

wseltzer: Have requested and are expecting to have availble

<hhalpin> So, a dial-in is likely and will be emailed to the mailing list once available.

Group Life

virginie: Listed topics for the F2F agenda
... Currently have two days laid out, but need to reclarify what was expected to have happen

hhalpin: Close all major issues on the first day
... Look at use cses on the second day
... If open issues bled intot he second day, that is more important

virginie: Need to look at test as well.
... Edit the Wiki to add things related to the current specifications if they are missing
... Agree main objective is to close open issues

<scribe> ... New issues will only be discussed if they are really new

UNKNOWN_SPEAKER: Want to freeze time for remote attendees to be able to get in
... Any europe based remote attendees?

nvdbleek: I think i might

virginie: Will make proposal on timings, let me know if that does not work well.
... Should come out next week
... Any other issues?

<hhalpin> If we have some time, we can kill issues :)

<hhalpin> Which group MichaelH?

<hhalpin> Chair likely forgot

<MichaelH> SysApps

virginie: Web security interset group had charter renewed. I am co-chairing and the gorup is now alive
... Looking at mobile web app securtiy model
... People in this group might be interesed in joining and looking at the mailing list
... Contact me or wendy to get more info

<MichaelH> @hhalpin so do I need to ask again somewhere?

<rsleevi> @MichaelH: email the chair of that group

<MichaelH> SysApps

<wseltzer> [adjourned]

<wseltzer> trackbot, end teleconf

Summary of Action Items

[NEW] ACTION: rsleevi to add note re: streams [recorded in http://www.w3.org/2013/10/28-crypto-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.138 (CVS log)
$Date: 2013-10-28 20:46:38 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.138  of Date: 2013-04-25 13:59:11  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: i/Scribe:/Topic: WebCrypto API
Succeeded: s/IDO/IDL/
Succeeded: s/markw/MichaelH/
Succeeded: s/including in the document/studying the proposal in detail/
Succeeded: s/???/arunranga:/
Succeeded: s/requerst/request/
Found ScribeNick: jimsch
Found Scribe: Jim Schaad
Default Present: Michael_Hutchinson, +1.617.253.aaaa, Virginie_Galindo, Wendy, israelh, Sangrae, markw, jimsch, jyates, arunranga, +1.650.275.aabb, rsleevi, +1.703.284.aacc, [Microsoft], rbarnes, vgb, nvdbleek, hhalpin
Present: Michael_Hutchinson +1.617.253.aaaa Virginie_Galindo Wendy israelh Sangrae markw jimsch jyates arunranga +1.650.275.aabb rsleevi +1.703.284.aacc [Microsoft] rbarnes vgb nvdbleek hhalpin

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 28 Oct 2013
Guessing minutes URL: http://www.w3.org/2013/10/28-crypto-minutes.html
People with action items: rsleevi

[End of scribe.perl diagnostic output]