[widgets] Minutes of 28 August 2008 Widgets f2f meeting

The minutes from the August 28 Widgets f2f meeting are available at  
the following and copied below:

  <http://www.w3.org/2008/08/28-wam-minutes.html>

WG Members - if you have any comments, corrections, etc., please send  
them to the public-webapps mail list before September 11 (next Widets  
voice conference); otherwise these minutes will be considered approved.

-Regards, Art Barstow


    [1]W3C

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

                                - DRAFT -

                           Widgets F2F Meeting
                               28 Aug 2008

    [2]Agenda

       [2] http://www.w3.org/2006/appformats/group/TurinF2F

    See also: [3]IRC log

       [3] http://www.w3.org/2008/08/28-wam-irc

Attendees

    Present
           Art_Barstow, Marcos_Caceres, Nick_Allot, David_Rogers,
           Mark_Priestly, Benoit_Suzzane, Claudio_Venezia, Dino_Gallo,
           Diego_Gibellino, Luca_Bruera, Maruo_Sacco, Mike_Smith

    Regrets
    Chair
           Art

    Scribe
           Art

Contents

      * [4]Topics
          1. [5]Agenda Review for Today
          2. [6]Issue #18 - Need to define a mechanism to check for the
             availability of an API
          3. [7]Issue #35 - SVG as an icon format
          4. [8]Issue #36 - Is the file API going to be part of Widgets
             1.0?
          5. [9]Issue #45 - Do we need an extensible metadata hook?
          6. [10]Issue #46 - Need to define a <span> for i18n purposes
             in configuration document
          7. [11]OMTP Security
          8. [12]WAF Action #182 - Contact the CAs regarding the reuse
             of TLS certs for Widgets
          9. [13]Landscape doc
         10. [14]Requirements Doc
         11. [15]Auto Updates
         12. [16]Packaging and Configuration spec
         13. [17]API and Event spec
         14. [18]Digital Signature spec
         15. [19]Mandelieu F2F Meeting
         16. [20]Implementations
         17. [21]Any Other Business
         18. [22]Thanks Claudio and Telecom Italia
      * [23]Summary of Action Items
      _________________________________________________________


    RRSAgenet, make log member

    <Benoit> morning

    Date: 28 August 2008

    <scribe> Scribe: Art

    <scribe> ScribeNick: ArtB

    RRSAgent make minutes

Agenda Review for Today

    AB: agenda is: [24]http://www.w3.org/2006/appformats/group/TurinF2F
    ... we continue discussions on the P&C spec in particular open
    Issues for that spec
    ... we can then continue any security or sig related discussions we
    want to have
    ... Nick has agreed to make a presentation about OMTP's relevant
    security work
    ... Lastly, we will talk about schedule and plans between now and
    Mandelieu

      [24] http://www.w3.org/2006/appformats/group/TurinF2F

Issue #18 - Need to define a mechanism to check for the availability of
an API

    AB: the issue is: [25]http://www.w3.org/2008/webapps/track/issues/18

      [25] http://www.w3.org/2008/webapps/track/issues/18

    MC: I submitted a proposal to address this issue
    ... we discussed it yesterday
    ... I propose to close this issue since it is captured in the latest
    ED for the API and Events spec

    AB: any objections to close this?

    MP: VF is ok with the proposal we discussed yesterday
    ... We do need to feed in a new use case or two

    BS: I'm OK with the proposal we discussed yesterday

    <MikeSmith> to ArtB: a request: If you could get the phone bridge on
    for Nick's presentation at least, and get a mic close to him while
    he's speaking, that would be great

    RESOLUTION: we will close Issue #18 and related discussions about
    the model will continue on the public mail list

    <scribe> ACTION: Barstow close Issue #18 with the resolution and
    rationale in the 28 Aug 2008 minutes [recorded in
    [26]http://www.w3.org/2008/08/28-wam-minutes.html#action01]

    <trackbot> Created ACTION-234 - Close Issue #18 with the resolution
    and rationale in the 28 Aug 2008 minutes [on Arthur Barstow - due
    2008-09-04].

    <Benoit> big table and a router's fan near the phone area... sorry

    <scribe> ACTION: David work with OMTP members to provide input on
    the enabling access to proprietary APIs model [recorded in
    [27]http://www.w3.org/2008/08/28-wam-minutes.html#action02]

    <trackbot> Sorry, amibiguous username (more than one match) - David

    <trackbot> Try using a different identifier, such as family name or
    username (eg. dorchard, drogers)

    <scribe> ACTION: Rogers work with OMTP members to provide input on
    the enabling access to proprietary APIs model [recorded in
    [28]http://www.w3.org/2008/08/28-wam-minutes.html#action03]

    <trackbot> Created ACTION-235 - Work with OMTP members to provide
    input on the enabling access to proprietary APIs model [on David
    Rogers - due 2008-09-04].

Issue #35 - SVG as an icon format

    AB: the issue is: [29]http://www.w3.org/2008/webapps/track/issues/35

      [29] http://www.w3.org/2008/webapps/track/issues/35

    MC: if a SVG image can take live events, how do we deal with it?
    ... Do we want to deal with it at all?
    ... What is the current state of support in the mobile world?

    BS: what is the status of SVG impl in the mobile space?

    Dino: there are some impls of SVG1.2 Tiny
    ... At least two of the impls are in mobiles
    ... The SVG spec includes the micro-DOM support and an event model
    ... I understand the issue but it could be a lot of work for a
    mobile impl
    ... There could be some room to create a profile.

    MC: my gut feel is not to create a profile
    ... May need to say something like "if you want to use an SVG icon,
    use SVG 1.2 Tiny"

    Dino: but may want to include some restrictions

    <scribe> ACTION: Marcos add SVG 1.2 Tiny as an icon format (to the
    P&C) spec and then ask the SVG WG for comments [recorded in
    [30]http://www.w3.org/2008/08/28-wam-minutes.html#action04]

    <trackbot> Created ACTION-236 - Add SVG 1.2 Tiny as an icon format
    (to the P&C) spec and then ask the SVG WG for comments [on Marcos
    Caceres - due 2008-09-04].

    AB: do we close this issue then?

    MC: yes

    AB: any objections to closing this issue?

    [ None ]

    RESOLUTION: Issue #35 is closed; SVG1.2 Tiny will be added to list
    of supported formats in the P&C spec

    <scribe> ACTION: Barstow close Issue #35 with the rationale above
    [recorded in
    [31]http://www.w3.org/2008/08/28-wam-minutes.html#action05]

    <trackbot> Created ACTION-237 - Close Issue #35 with the rationale
    above [on Arthur Barstow - due 2008-09-04].

Issue #36 - Is the file API going to be part of Widgets 1.0?

    AB: issue #36 is: [32]http://www.w3.org/2008/webapps/track/issues/36

      [32] http://www.w3.org/2008/webapps/track/issues/36

    <MikeSmith> ArtB: conference Team_(MikeSmith)08:53Z scheduled with
    code 26633 (CONF3) for 60 minutes until 0953Z

    MC: Opera proposed a new file access API for Widgets last Spring
    ... Arve doesn't think it should be part of the "core" Widget API
    ... does OMTP have a need for this?

    NA: yes, something like that is in scope for us

    <claudio> TI's SVG guy is Diego not Dino

    NA: If it isn't part of the Core, where would it be defined?

    <MikeSmith> ArtB: OK, you can dial into Zakim at any time

    MC: it would be a separate spec created by WebApps WG

    <MikeSmith> tlr: I think David Rogers will be doing a presentation
    about OMTP security shortly

    AB: I prefer a smallish core and then some extensions

    NA: is the extensibility mechanism explicit?

    MC: yes, the extensibility model will be part of the core
    ... Timing wise, the core and other APIs can proceed separately but
    they could also be synch'ed up provided an appropriate level of
    staffing
    ... we need an Editor for the File API

    <tlr> mike, thanks for the ping; on a call now

    NA: I can't make any commitments but I can look into it

    MC: we need competent Editors that understand the relative urgency
    to complete our specs
    ... what is the process for WebApps starting new APIs?

    AB: the Charter addresses this issue
    ... In general, if there is a new API, we need to get AC approval
    before we start
    ... so where are we on this issue?

    MC: I don't think File should be a core API

    AB: propose that File API not be considered part of the Widgets API
    Core
    ... any objections?

    BS: does this mean a new doc will be created?

    MC: yes that is the expectation

    AB: we need someone to take ownership

    <scribe> ACTION: Barstow work with Nick and Charles to find an
    Editor for the File API spec [recorded in
    [33]http://www.w3.org/2008/08/28-wam-minutes.html#action06]

    <trackbot> Created ACTION-238 - Work with Nick and Charles to find
    an Editor for the File API spec [on Arthur Barstow - due
    2008-09-04].

    [ No objections to the proposal above re #36 ]

    RESOLUTION: Issue #36 is Closed via the rationale above

    <scribe> ACTION: Barstow close Issue #36 with the rational above
    [recorded in
    [34]http://www.w3.org/2008/08/28-wam-minutes.html#action07]

    <trackbot> Created ACTION-239 - Close Issue #36 with the rational
    above [on Arthur Barstow - due 2008-09-04].

Issue #45 - Do we need an extensible metadata hook?

    AB: the issue is: [35]http://www.w3.org/2008/webapps/track/issues/45
    ... what does this "metadata extension" mechanism really mean?

      [35] http://www.w3.org/2008/webapps/track/issues/45

    MC: he basically wants an RDF model

    BS: or is he saying the packaging format should not break if it
    contains unknown elements

    MC: I think we need to wait for the market to demand the need for
    additional metadata

    CV: could look at semantic annotation for XML Schema

    MC: I don't want to add such a dependency
    ... we already have an extension mechanism -> XML Namespaces
    ... and then the Author can add anything they want
    ... Our processing model explicitly says to ignore unknown elements
    and attributes

    CV: what about use cases for discovery?
    ... adding some additional semantics would be good

    AB: I agree adding more semantics would be good but I think our
    current model supports that

    BS: so we can close this issue right?

    MC: yes.
    ... Propose to close #45 because we already provide an extension
    mechanism to add additional metadata to a manifest.

    AB: any objections to that proposal?

    [ None ]

    RESOLUTION: Issue #45 will be closed - we already provide an
    extension mechanism to add additional metadata to a manifest

    <scribe> ACTION: Barstow close Issue #45 with the rationale above
    [recorded in
    [36]http://www.w3.org/2008/08/28-wam-minutes.html#action08]

    <trackbot> Created ACTION-240 - Close Issue #45 with the rationale
    above [on Arthur Barstow - due 2008-09-04].

    <scribe> ACTION: Claudio add extensible metadata model for the
    manifest to the v2 feature list [recorded in
    [37]http://www.w3.org/2008/08/28-wam-minutes.html#action09]

    <trackbot> Created ACTION-241 - Add extensible metadata model for
    the manifest to the v2 feature list [on Claudio Venezia - due
    2008-09-04].

    <scribe> ACTION: Barstow send an e-mail to public-appformats that
    enumerates the Issues we closed this week and includes the rationale
    [recorded in
    [38]http://www.w3.org/2008/08/28-wam-minutes.html#action10]

    <trackbot> Created ACTION-242 - Send an e-mail to public-appformats
    that enumerates the Issues we closed this week and includes the
    rationale [on Arthur Barstow - due 2008-09-04].

Issue #46 - Need to define a <span> for i18n purposes in configuration
document

    AB: we discussed this on Aug 26 but we didn't assign any actions

    <scribe> ACTION: Marcos ask I18N WG if Unicode RTL is sufficiently
    supported in UAs; if not we will include the <span> element in the
    Widgets spec [recorded in
    [39]http://www.w3.org/2008/08/28-wam-minutes.html#action11]

    <trackbot> Created ACTION-243 - Ask I18N WG if Unicode RTL is
    sufficiently supported in UAs; if not we will include the <span>
    element in the Widgets spec [on Marcos Caceres - due 2008-09-04].

OMTP Security

    <MikeSmith> tlr: Nick getting started now

    NA: after I cleanup the slide I am presenting, I will send a copy to
    public-webapps
    ... OMTP is mainly a requirement group
    ... We have done a lot of reqs related to security fwks
    ... Some of our work is relevant to Widget UAs
    ... A lot of work is on the security policy framework
    ... BONDI is a "different" project for OMTP because
    ... instead of just reqs, we expect to create a Reference
    Implementation (RI)
    ... The area of "enhanced web runtimes" is of wide interest in the
    industry
    ... We see concerns about fragmentation in this space, especially
    regarding API fragmenation

    [ Nick show block diagram of the Architecture ]

    NA: a key part of the sec fwk is identity and we have some different
    models re identify e.g. certs
    ... all of these identity models are in scope
    ... we expect the policy fwk to cover these various models
    ... regarding our APIs of interest, we have to deal with a) Generic
    Event Mechanism
    ... b) JavaScript Errors

    MC: are the Web Package and Widget Package blocks different?

    NA: yes, they could be e.g. they could have different identity
    associated with them
    ... Some could have signatures; some not ...
    ... Want a clear seperation of the application identity and
    application authorization
    ... This model will be declarative in a policy file
    ... We think our model will be much more flexible
    ... One underlying assumption is to minimize user interaction re
    security considerations, policies, etc.
    ... Our fwk is agnostic as to business models
    ... There can be a policy that separates Widgets into two groups:
    ones that have no privs; ones that have lots of privs

    CV: is this similar to a black/white list model, Marcos?

    MC: it's similar but more complicated

    MP: there can be diffs between policies and white/black lists

    CV: what will the policy language include?

    NA: we want a language that will support a wide range of policies

    AB: will you create your own policy language?

    NA: we will use OASIS' XACML
    ... Fabio is defining a dictionary mapping for us
    ... It could be XACML is too heavy-weight for some devices in scope
    for us

    AB: has OASIS done some profileing of XACML?

    Fabio: we need to identify a subset; we are discussing a general fwk
    ... we may identify some profiles
    ... we are still working through some scenarios

    NA: security policies can be very complicated
    ... and they can affect the user experience
    ... must also reflect user's specific preferences
    ... must also respect user's privacy requirements and some
    jurisdictions have legal frameworks that must be adhered to
    ... As to the APIs, we have about 10 that are of interest to us
    ... Like WebApps' Widgets work, we recognize a need for an
    extensible API model
    ... But this model must not break the security model
    ... The APIs are:
    ... Application Settings - can be app-specific or shared settings
    ... User Interaction, Location, PIM, Phone Status, Persistence,
    Gallery, Messaging, Application Invoke, Telephony, Camera

    s/Applicatin, Invoke,/Application Invoke,/

    NA: Gallery API is for an app to access all multi-media on a device
    ... re Persistence, we could just re-use the Opera proposal
    ... re Location API, we'll probably use or re-use the GeoLoca work
    being done at the W3C

    MC: are you talking to the GeoLoc WG?

    AB: I don't think that WG has been Chartered yet

    MS: I expect an annoucement soon-ish re the GeoLoc WG
    ... Matt will be the Team Contact

    NA: we may be able to use the DCCI fwk
    ... but no hard decision has been made yet regarding DCCI

    AB: it is my understanding the OpenAjax Alliance has these APIs in
    scope too

    NA: conceptually, these APIs are in scope for them but I don't think
    they've done much

    MP: OAA has a security group and they passed that work to OMTP
    ... Regarding APIs, OAA is interested in just a shim layer on top of
    "real" APIs

    NA: we have a comm channel with OAA and will keep it going

    AB: I would to understand more about the expectations for the RIs

    NA: we expect contributions from OMTP members
    ... the RIs may not result in re-usable code
    ... We are keeping the licensing terms as flexible as possible
    ... Expect some to be GPL or GPL-like; we also expect some binary
    components

    AB: are you working at all with the UWA WG?

    NA: I've talked with the Chair and Team Contacts
    ... No formal agreements as of now regarding how to cooperate
    ... of the ten APIs we've identified, what are the mappings to W3C
    and other Standard Orgs

    CV: does OMTP have a relationship with the MWI?

    DR: we are investigating it; no firm decisions yet
    ... we (OMTP) are resource-constrained

WAF Action #182 - Contact the CAs regarding the reuse of TLS certs for
Widgets

    MP: what is the issue here?

    AB: I'm not exactly sure
    ... I'll need to talk to Thomas

    MP: my question is: is the desire to use TLS certs to sign a Widget
    package?

    <mpriestl> Concern is that TLS certs are not used to sign widget
    packages. Certificates are issued based on the presumption that they
    will be used for a specific purpose. We would object to bending
    these rules for widgets. If the desire is to use TLS certs for TLS
    then this would obviously be fine! Request that reason for question
    is clarified.

    AB: OK then, I propose we close this Action

    MC: I agree

    AB: any objections to closing Action 182?

    [ None ]

    AB: Note WAF Action #182 is the same as WebApps Action #206
    ([40]http://www.w3.org/2008/webapps/track/actions/206)

      [40] http://www.w3.org/2008/webapps/track/actions/206)

Landscape doc

    AB: what's the plan, Marcos?

    MC: I only plan to make minor edits

    AB: any issues or concerns?

    BS: is Webwag a candidate?

    MC: it's a closed system

    AB: without any commitments for contribution, I would not worry
    about

    MC: I will complete it some day because it is part of my PhD and
    that means I will be done by the end of December.
    ... I want to finish it as a WG Note

    AB: excellent; it's been a very valuable resource!

    MC: I would also like to be official "Author" of that doc

    AB: I support that

    BS: me too

Requirements Doc

    MC: next is to process the OMTP input
    ... I think we will need to go back to Working Draft
    ... And then do a minimal-length LC
    ... I want that comment period to end before October 20

    AB: do we need to publish a new WD before we publish a new LC WD
    ... can our next pub be another LC?

    MS: yes, we can do that
    ... any number of LCs is possible and any number in a row is
    possible, I think

    AB: so the plan is to complete the OMTP review within the next 2-3
    weeks and to be ready to submit for publication by roughly Sept 20

    BS: but we want the comment period to end one week before we meet in
    Mandelieu

    MC: I will try to have it ready to publish by Sep 12th

    AB: excellent, Marcos!

Auto Updates

    MC: I want a FPWD on September 19

    DR: we have a problem with the Reqs proposal

    MC: I will publish the Reqs doc on September 19 and we will ask for
    a 3-week review period
    ... that will give us one week to review the comments before our
    Mandelieu meeting

    AB: is that OK with you David?

    DR: yes

    NA: yes

    AB: back to Auto Updates ...

    MC: I will shoot for a September 12 FPWD
    ... can OMTP guys live with that date

    AB: this would mean that during our Sep 11 Voice Conf we should
    record a "consensus" decision to publish this FPWD

    MC: people can start looking at the latest ED now; I don't expect a
    lot of changes

Packaging and Configuration spec

    MC: I propose the next pub on October 3
    ... it will be another WD
    ... Wait, Wait, it will be ready for member review on Oct 3
    ... My expectation for Mandelieu is: after a short (1-2 weeks)
    period after the meeting, we should be ready to publish a LC WD

    AB: sounds like a good plan

    MC: On October 31, I plan to submit it the webreq team for
    publication as a LC WD.

    <scribe> ACTION: try to schedule some f2f time with the TAG during
    Mandelieu re the widget: scheme issue [recorded in
    [41]http://www.w3.org/2008/08/28-wam-minutes.html#action12]

    <trackbot> Sorry, couldn't find user - try

    <MikeSmith> trackbot, status

    <trackbot> This channel is not configured

    <MikeSmith> trackbot, status?

    <trackbot> This channel is not configured

    <scribe> ACTION: Barstow try to schedule some f2f time with the TAG
    during Mandelieu re the widget: scheme issue [recorded in
    [42]http://www.w3.org/2008/08/28-wam-minutes.html#action13]

    <trackbot> Created ACTION-244 - try to schedule some f2f time with
    the TAG during Mandelieu re the widget: scheme issue [on Arthur
    Barstow - due 2008-09-04].

    BS: because of the TP blackout, may not be able to publish on Oct 31
    but a week later at least
    ... this would mean the earliest we would exit LC is approx December
    1

    <MikeSmith> [[

    <MikeSmith> 13 October, 12pm ET: Deadline for publication requests

    <MikeSmith> before moratorium

    <MikeSmith> 16 October: Last publications before moratorium

    <MikeSmith> 17 - 26 October: No publications

    <MikeSmith> 27 October: Publications resume

    <MikeSmith> ]]

API and Event spec

    MC: Arve said his target for FPWD is mid-September

    AB: do you have any concerns about that Marcos?

    MC: no

    AB: so tentatively, we would be ready to make a formal decision re
    the FPWD during our September 18 Voice Conf

    CV: did we get consensus on the title of the spec

    MC: I want to talk to Arve about that

Digital Signature spec

    AB: what are our plans for the DigSig spec?

    MC: hope to have something ready for the TP
    ... Plan a new ED to discuss by October 17
    ... I'll try earlier but I can't guarantee anything

    AB: can Marcos get some help on that spec?

    MC: I'm planning to work with Mark and David
    ... From October 6-15 I will focus on that spec

    AB: I will start dialog with XML Sec WG to see if they can provide
    some input (and not just review)

Mandelieu F2F Meeting

    CV: are the dates confirmed?

    AB: yes, Oct 20 and 21
    ... I will submit a detailed agenda at least two weeks before the
    meeting

Implementations

    CV: does anyone have any plans they can disclose?

    MC: I started a RI but I had to stop it because of all of the
    editing work I'm doing

    Fabio: what is you RI?

    MC: it's a JS impl that codifies every assertion in the spec

    Fabio: perhaps there could be some cooperation with OMTP on the RI

Any Other Business

    MC: David, when is OMTP going to bring over the API specs?

    NA: regarding reqs, that stuff is available now on our Web site
    ... We cannot submit it to the W3C until the IPR issues are resolved
    ... The details of our specs are tied to our RIs

    <drogersuk> We will discuss further in Austin

Thanks Claudio and Telecom Italia

    AB: thanks very much for hosting us Claudio!
    ... The food, drink and everything!
    ... Meeting Adjourned

Summary of Action Items

    [NEW] ACTION: Barstow try to schedule some f2f time with the TAG
    during Mandelieu re the widget: scheme issue [recorded in
    [43]http://www.w3.org/2008/08/28-wam-minutes.html#action13]
    [NEW] ACTION: Barstow close Issue #18 with the resolution and
    rationale in the 28 Aug 2008 minutes [recorded in
    [44]http://www.w3.org/2008/08/28-wam-minutes.html#action01]
    [NEW] ACTION: Barstow close Issue #35 with the rationale above
    [recorded in
    [45]http://www.w3.org/2008/08/28-wam-minutes.html#action05]
    [NEW] ACTION: Barstow close Issue #36 with the rational above
    [recorded in
    [46]http://www.w3.org/2008/08/28-wam-minutes.html#action07]
    [NEW] ACTION: Barstow close Issue #45 with the rationale above
    [recorded in
    [47]http://www.w3.org/2008/08/28-wam-minutes.html#action08]
    [NEW] ACTION: Barstow send an e-mail to public-appformats that
    enumerates the Issues we closed this week and includes the rationale
    [recorded in
    [48]http://www.w3.org/2008/08/28-wam-minutes.html#action10]
    [NEW] ACTION: Barstow work with Nick and Charles to find an Editor
    for the File API spec [recorded in
    [49]http://www.w3.org/2008/08/28-wam-minutes.html#action06]
    [NEW] ACTION: Claudio add extensible metadata model for the manifest
    to the v2 feature list [recorded in
    [50]http://www.w3.org/2008/08/28-wam-minutes.html#action09]
    [NEW] ACTION: David work with OMTP members to provide input on the
    enabling access to proprietary APIs model [recorded in
    [51]http://www.w3.org/2008/08/28-wam-minutes.html#action02]
    [NEW] ACTION: Marcos add SVG 1.2 Tiny as an icon format (to the P&C)
    spec and then ask the SVG WG for comments [recorded in
    [52]http://www.w3.org/2008/08/28-wam-minutes.html#action04]
    [NEW] ACTION: Marcos ask I18N WG if Unicode RTL is sufficiently
    supported in UAs; if not we will include the <span> element in the
    Widgets spec [recorded in
    [53]http://www.w3.org/2008/08/28-wam-minutes.html#action11]
    [NEW] ACTION: Rogers work with OMTP members to provide input on the
    enabling access to proprietary APIs model [recorded in
    [54]http://www.w3.org/2008/08/28-wam-minutes.html#action03]
    [NEW] ACTION: try to schedule some f2f time with the TAG during
    Mandelieu re the widget: scheme issue [recorded in
    [55]http://www.w3.org/2008/08/28-wam-minutes.html#action12]

    [End of minutes]
  

Received on Saturday, 30 August 2008 13:18:39 UTC