Draft minutes 2010-05-26

Draft minutes  from teleconference on 2010-05-26 for approval at next meeting. Thanks Daniel for scribing. HTML follows text.

   [1]W3C

      [1] http://www.w3.org/

                               - DRAFT -

          Device APIs and Policy Working Group Teleconference

26 May 2010

   [2]Agenda

      [2] http://lists.w3.org/Archives/Public/public-device-apis/2010May/0123.html

   See also: [3]IRC log

      [3] http://www.w3.org/2010/05/26-dap-irc

Attendees

   Present
          alissa, Dom, +1.314.454.aaaa, enewland, darobin, Dzung_Tran,
          Robin_Berjon, Dominique_Hazael-Massieux, Eric, Newland,
          Alissa_Cooper, LauraA, Marco_Marengo, Daniel_Coloma,
          Claes_Nilsson, Richard_Tibbett, Anssi_Kostiainen,
          Wojciech_Maslowski, Wojciech_Masłowski, Suresh_Chitturi

   Regrets
          Frederick_Hirsch, John_Morris, Arve_Bersvendsen, Frederick

   Chair
          Robin_Berjon

   Scribe
          Daniel

Contents

     * [4]Topics
         1. [5]announcements
         2. [6]Minutes Approval
         3. [7]Policy Segment
         4. [8]APIs
         5. [9]General API Review
     * [10]Summary of Action Items
     _________________________________________________________

   <trackbot> Date: 26 May 2010

   <arve> I have conflicting appointments

   <darobin> stupid bot

   <darobin> okay let's wait a little

   <darobin> danielcoloma, can you scribe? you're at the top of the
   list

   sure

   <darobin> thanks!

   <darobin> Scribe: Daniel

   <darobin> ScribeNick: danielcoloma

announcements

   <dom> [11]Registration for DAP F2F in London, July 2010

     [11] http://www.w3.org/2002/09/wbs/43696/london2010/

   dom: registration for London meeting is open

   <dom> [12]Current registrants for F2F

     [12] http://www.w3.org/2002/09/wbs/43696/london2010/results

   <darobin>

   darobin: privacy workshop is scheduled for the same week as DAP F2F

Minutes Approval

   <darobin> [13]Minutes of 19 May 2010 call

     [13] http://lists.w3.org/Archives/Public/public-device-apis/2010May/att-0093/minutes-2010-05-19.html

   RESOLUTION: minutes approved

Policy Segment

   <dom> ACTION-152?

   <trackbot> ACTION-152 -- Laura Arribas to edit policy framework,
   reviewing BONDI material and editorial update -- due 2010-05-05 --
   OPEN

   <trackbot> [14]http://www.w3.org/2009/dap/track/actions/152

     [14] http://www.w3.org/2009/dap/track/actions/152

   <dom> [15]Laura's report on progress on policy framework wrt trust
   domains

     [15] http://lists.w3.org/Archives/Public/public-device-apis/2010May/0127.html

   LauraA: Changes submitted to CVS this morning
   ... e-mail summarizes key changes done in the commit

   <darobin> [16]the Policy document

     [16] http://dev.w3.org/2009/dap/policy/

   LauraA: Trust domain control layer added, new dataflow

   <dom> [17]Dataflow diagram

     [17] http://dev.w3.org/2009/dap/policy/dataflow.png

   LauraA: One of the objectives was making it simpler, not sure if it
   is achieved yet
   ... need help to split the document in different specs

   darobin: volunteers for contributing more to the work?

   <wojciech> yes it's me

   darobin: which kind of problems do you have when splitting the doc?

   lauraA: agreed to have 3 docs (policy framework, profile and example
   documents) but not sure how to do the split

   darobin: create 3 copies of the doc, add them to the CVS and remove
   the part that is not needed for every of them
   ... Do you think the documents resulting of the split are
   small/indepent enough to be implemented in a granular manner?

   LauraA: Yes, I think so

   <AnssiK> editorial comment re policy framework doc: do not scale
   images, instead display them in their native size for better
   readability (see e.g. section 2.3)

   darobin: status from a publication point view?

   LauraA: The policy framework is the more mature one

   darobin: could we have a FPWD Cfc next week?

   LauraA: Yes, it is doable

APIs

   darobin: discussion on the sysinfo attributes on the mailing list
   ... target is reach an agreement today
   ... MAC address and SSID, suggestion is remove both

   <dom> [18]Analysis of SysInfo Attributes

     [18] http://lists.w3.org/Archives/Public/public-device-apis/2010May/0098.html

   <Dzung_Tran> +1

   <maxf1> +1

   dom: widget and open web are different use cases

   darobin: we should think about the use case for exposing these
   attributes

   <tlr> I like Robin's proposal.

   <dom> (this could be something defined in the policy framework, that
   said)

   darobin: accessing some properties may return a denied permission
   error depending on whether an installed applciation or a web page is
   accessing them

   <Dzung_Tran> At one point we talk about a different use cases
   document, Is there one?

   darobin: the other alternative is having an extended spec for these
   additional properties

   tlr: supports the idea of having these properties in another spec

   <alissa> +1 to having scary properties separate

   dom: who is doing that analysis?

   <Dzung_Tran> that analysis would be difficult, I would just remove
   obvious scary properties for release 1.0, revisit in release 2.0

   <dom> (an editors note in the doc maybe?)

   dom: We could include a note saying that these properties are not
   part of the spec because of privacy issues

   <darobin> PROPOSED RESOLUTION: macAddress and SSID removed; Make a
   note that these have been removed but that the group is considering
   making them available in a more advanced document

   <darobin> + because of privacy concerns

   <darobin> + seeking comments, planning to revisit in future version
   => in SotDF

   <darobin> RESOLUTION: macAddress and SSID removed; Make a note in
   SotD that these have been removed but that the group is seeking
   comments and considering making them available in a future version

   <darobin> ACTION: maxf to implement the maxAddress/SSID resolution
   [recorded in
   [19]http://www.w3.org/2010/05/26-dap-minutes.html#action01]

   <trackbot> Created ACTION-175 - Implement the maxAddress/SSID
   resolution [on Max Froumentin - due 2010-06-02].

   darobin: suggestion to remove IP address

   <tlr> ack

   <darobin> PROPOSED RESOLUTION: remove ipAddress as well, put in same
   list as macAddress and SSID

   <darobin> RESOLUTION: remove ipAddress as well, put in same list as
   macAddress and SSID

   darobin: APN, operatorName, roaming, mcc and mnc have been also
   suggested for being removed
   ... suggest to separate roaming from the other four, as there are
   strong use cases for it at least

   <tlr> The "data connection is expensive" flag, right.

   alissa: some of these attributes may be redundant

   maxf: Bryan was the main supporter for including these attributes,
   we may need to check with him

   darobin: suggestion is removing them unless we have strong use cases
   for them

   tlr: it would be good to validate the decission on the mailing list

   <darobin> ACTION: Robin to email Bryan to ask what his use cases are
   for apn, operatorName, mcc, mnc [recorded in
   [20]http://www.w3.org/2010/05/26-dap-minutes.html#action02]

   <trackbot> Created ACTION-176 - Email Bryan to ask what his use
   cases are for apn, operatorName, mcc, mnc [on Robin Berjon - due
   2010-06-02].

   tlr: roaming is a way to abstract to the user he may be charged for
   data connections more expensively

   darobin: the use case is the "expensive flag", but not sure how to
   specify it in a way that is useful

   <darobin> ACTION: Robin to ask the mailing about usage of roaming
   [recorded in
   [21]http://www.w3.org/2010/05/26-dap-minutes.html#action03]

   <trackbot> Created ACTION-177 - Ask the mailing about usage of
   roaming [on Robin Berjon - due 2010-06-02].

   darobin: suggestion is keeping the rest of elements (included in
   item 4) in the spec

   <darobin> RESOLUTION: keep the fields: type,
   currentDownloadBandwidth, currentUploadBandwidth,
   maxDownloadBandwidth, maxUploadBandwidth, currentSignalStrength

   Suresh: From a conformance point of view, could we group some of
   these properties
   ... it should be possible to query if one property group is
   supported or not

   darobin: this approach has not worked in DOM, SVG

   <Zakim> dom, you wanted to note link with policy framework

   Suresh: Key question is the sysinfo spec a all or none spec?

   darobin: conformance is all or none, but for instance, if sensor
   does not return a value it would be acceptable

   <dom> [I think making a concrete proposal would help in any case :)
   ]

   darobin: if we want it to be modular, we should split the spec
   ... handling it at the conformance level is very difficult

   maxf: is there any conclusion on the issue of having multiple active
   connections?

   darobin: no resolution so far

   alissa: privacy discussions should be held when we have reached a
   conclusion on the multiple connections issue

   <darobin> [22]Suresh's comments

     [22] http://www.w3.org/mid/D37CC1B151BD57489F4F89609F168FE805838211@XCH01DFW.rim.net

   <maxf1> what's an "MMS connection"?

   Suresh: we need also to clarify the meaning of connections in this
   context

   <darobin> [I wonder if we don't have to expose multiple just to make
   sure we can be interoperable...]

   darobin: what are the downsides of having an array of multiple
   connections?
   ... complexity, interoperability problems, any other?

   <darobin> PROPOSED RESOLUTION: we only expose active connections and
   we expose all of them through activeConnections

   <darobin> RESOLUTION: we only expose active connections and we
   expose all of them through activeConnections

   <maxf1> ACTION maxf to make activeConnection multiple

   <trackbot> Created ACTION-178 - Make activeConnection multiple [on
   Max Froumentin - due 2010-06-02].

General API Review

   darobin: in general, the progress of the APIs has been slowed down
   ... which are the reasons for that?

   <dom> (tracker is supposed to be used for that; not that I mind
   using the wiki for it)

   richt: pending resolutions on some issues may be a reason

   <Dzung_Tran> I think the main reason got to do with privacy

   tlr: we already have the tracker

   darobin: kind of wiki in which we list the key issues for every API
   may help

   <darobin> ACTION: Robin to get the ball rolling on documenting path
   forward for specs [recorded in
   [23]http://www.w3.org/2010/05/26-dap-minutes.html#action04]

   <trackbot> Created ACTION-179 - Get the ball rolling on documenting
   path forward for specs [on Robin Berjon - due 2010-06-02].

   <maxf1> ah

   tlr: concerns on the community about the capture API

   darobin: still waiting for input from mozilla

   <wojciech> yes

   tlr: some concerns may be due to confusion
   ... suggestion is putting on the list a clarification on the scope
   of the capture API

   <richt> sorry I dropped from the call.

   <richt> I will document progress on the wiki for Contacts and
   Calendar

   <darobin> ACTION: Robin to ping the editors about explaining the
   design of Capture on the list [recorded in
   [24]http://www.w3.org/2010/05/26-dap-minutes.html#action05]

   <trackbot> Created ACTION-180 - Ping the editors about explaining
   the design of Capture on the list [on Robin Berjon - due
   2010-06-02].

   maxf: last call in this WG, passing the editorshp on sysinfo the
   Dzung_Tran

   <scribe> ... new Opera employee on this group, wojciech

   <Dzung_Tran> Max, wish you well

   <maxf1> thanks :)

   <dom> welcome wojciech!

   <darobin> RRSAgent: make minutes

Summary of Action Items

   [NEW] ACTION: maxf to implement the maxAddress/SSID resolution
   [recorded in
   [25]http://www.w3.org/2010/05/26-dap-minutes.html#action01]
   [NEW] ACTION: Robin to ask the mailing about usage of roaming
   [recorded in
   [26]http://www.w3.org/2010/05/26-dap-minutes.html#action03]
   [NEW] ACTION: Robin to email Bryan to ask what his use cases are for
   apn, operatorName, mcc, mnc [recorded in
   [27]http://www.w3.org/2010/05/26-dap-minutes.html#action02]
   [NEW] ACTION: Robin to get the ball rolling on documenting path
   forward for specs [recorded in
   [28]http://www.w3.org/2010/05/26-dap-minutes.html#action04]
   [NEW] ACTION: Robin to ping the editors about explaining the design
   of Capture on the list [recorded in
   [29]http://www.w3.org/2010/05/26-dap-minutes.html#action05]

   [End of minutes]