The draft minutes from the April 23 Widgets voice conference are
available at the following and copied below:
<http://www.w3.org/2009/04/23-wam-minutes.html>
WG Members - if you have any comments, corrections, etc., please
sendthem to the public-webapps mail list before 30 April 2009 (the
next Widgets voice conference); otherwise these minutes will be
considered Approved.
-Regards, Art Barstow
[1]W3C
[1] http://www.w3.org/
- DRAFT -
Widgets Voice Conference
23 Apr 2009
[2]Agenda
[2] http://lists.w3.org/Archives/Public/public-webapps/
2009AprJun/0258.html
See also: [3]IRC log
[3] http://www.w3.org/2009/04/23-wam-irc
Attendees
Present
Art, Frederick, David, Robin, Marcos, Mark, Arve, Andy,
Marcin, Andrew, Mike
Regrets
Chair
Art
Scribe
Art
Contents
* [4]Topics
1. [5]Review and tweak agenda
2. [6]Announcements
3. [7]DigSig: comments by Mark
4. [8]DigSig: ECDSA and v1
5. [9]DigSig: getting ready for Last Call
6. [10]P&C: Dropping screenshot
7. [11]P&C: <access> element
8. [12]P&C: Localization proposal from Marcos
9. [13]Window Modes spec: status and plans
* [14]Summary of Action Items
_________________________________________________________
<scribe> ScribeNick: ArtB
<scribe> Scribe: Art
Date: 23 April 2009
Review and tweak agenda
AB: we will drop 3a. and 3c. since consensus for both of these was
achieved via email after I posted the agenda. Will add a new agenda
item about ECDSA. Are there any other change requests?
[ None ]
Announcements
AB: Please remember to register for the London F2F meeting June 9-11
([15]http://www.w3.org/2002/09/wbs/42538/WidgetsLondonJune2009/).
... the first voice conference of the Widgets Updates PAG is
tentative scheduled for 13:00 Boston time on April 28
([16]http://lists.w3.org/Archives/Member/member-widgets-pag/2009Apr/
0002.html) but Rigo Wenning hasn't yet confirmed that call.
... any other annoucements?
[15] http://www.w3.org/2002/09/wbs/42538/WidgetsLondonJune2009/).
[16] http://lists.w3.org/Archives/Member/member-widgets-pag/
2009Apr/0002.html)
[ None ]
DigSig: comments by Mark
AB: on April 7 Mark submitted a relatively long list of comments
([17]http://lists.w3.org/Archives/Public/public-webapps/2009AprJun/0
070.html). Frederick and Marcos responded. If addressing any of
these comments could benefit from some discussion today, we can
allocate some time. Frederick, what's the status?
[17] http://lists.w3.org/Archives/Public/public-webapps/
2009AprJun/0070.html).
FH: I think we've reached consensus on most comments
... one issue is sig file
... I use "widget sig" rather than file
... I think both usages makes the most sense depending on context
MP: I agree most comments have been addressed
... but I still need to do some review
AB: let's drop this topic for today and take any followups on the
mail list
DigSig: ECDSA and v1
AB: the ECDSA issue (captured reasonably well as Issue #81
[18]http://www.w3.org/2008/webapps/track/issues/81) is still open.
The latest ED does two related things: 1) it includes a note that
requests feedback on ECDSA; 2) it also does NOT mandate ECDSA as one
of the Signature Algorithms (and thus is a departure from latest WD
of XML Signature 1.1). Today there was more discussion on this
([19]http://lists.w3.org/Archives/Public/public-webapps/2009AprJun/0
279.html). One q
[18] http://www.w3.org/2008/webapps/track/issues/81)
[19] http://lists.w3.org/Archives/Public/public-webapps/
2009AprJun/0279.html).
FH: I think we need some text about it
... we cannot ignore it
... but we don't know what XML Sig 1.1 will decide
... using SHOULD may be a good way forward
... their risks with other algorightms so having an alternative to
consider
<drogersuk> I agree with FJH
FH: I do agree MUST is too strong
DR: I agree with FH
<Marcos> +q
DR: failure to indicate a roadmap is not a good path to take
MC: I understand the concerns here
... if company X implements all of the algs except ECDSA and then
they get a widget with ECDSA, there is a prob
... if SHOULD is used it will effectively make it required for the
implementors
... it is expensive to implement to implement all of these algs
... perhaps for v1.1. we could add support for new algs
... would like some real proof there is real market demand for EC
... but without that evidence I don't support including it
... think it should either be MUST or not in the spec at all
DR: I understand your concerns Marcos
... need to try to forsee future issues with a limited set
... VF have made it clear they see it on their roadmap
FH: I understand the issues with switching suites
... we are following this in XML Sec WG
... we see a demand from US gov't at least
MC: OK, that's good information
... and in that case, perhaps it should be a MUST
FH: in XML Sec WG we are leaning toward a MUST but are considering
various concerns
... think SHOULD would be a good indicator
AB: agree SHOULD would be a reasonable compromise
... we need to make it clear we want feedback from Implementors as
well as Developers
MP: as I said on the list list, we think SHOULD is the best thing
for now
... we can't use MUST at this point in time
MC: so let's go with SHOULD
AB: any additional comments?
[ None ]
<fjh> ECDSAwithSHA256
AB: propose a resolution: we will add ECDSA support as a SHOULD in
the Widgets DigSig spec
<fjh> proposed resolution - add ECDSAwithSHA256 as should in widgets
digsig
AB: any objections to my proposal?
[ None ]
RESOLUTION: we will add ECDSAwithSHA256 as a SHOULD in the Widgets
DigSig spec
DigSig: getting ready for Last Call
AB: the basic question is what, specifically, needs to be done
before this spec is "feature-complete" and hence for Last Call WD
publication? Frederick responded to this yesterday
([20]http://lists.w3.org/Archives/Public/public-webapps/2009AprJun/0
261.html).
[20] http://lists.w3.org/Archives/Public/public-webapps/
2009AprJun/0261.html).
FH: one question is how to work this with Sig Properties spec
... not sure how to work this out as a Chair
... maybe we take this offline
AB: my gut feel is your WG should publish Sig Properties
FH: we probably want a single annoucement
<scribe> ACTION: Barstow work with Frederick re synching Signature
Properties spec with Widgets DigSig spec [recorded in
[21]http://www.w3.org/2009/04/23-wam-minutes.html#action01]
<trackbot> Created ACTION-335 - Work with Frederick re synching
Signature Properties spec with Widgets DigSig spec [on Arthur
Barstow - due 2009-04-30].
<fjh> agree, xml security should publish signature properties. need
to coordinate however.
AB: are we Feature Complete?
FH: I don't think the reqs are in synch with the Reqs doc
MC: the problem is the links point to the ED rather then what is
/TR/
... we fix this when we publish these two docs in /TR/
AB: my recommendation is that when we publish the LCWD of DigSig we
also at the same time publish a new Reqs doc
... are we done with functionality?
FH: I think yes
<mpriestl> +1
MC: yes
MP: yes
<darobin> +1 on top of mpriestl
RB: yes
MP: yes, just need to update ECDSA
<fjh> need to add ECDSA, possibly other minor editorial tweaks
AB: propose Resolution: the Widgets Digital Signature spec is
Feature Complete; we will not add any new functionality
... any comments on that proposal?
... any objections?
[ None ]
DR: how would this affect schedule?
MC: I think this would put us ahead of scheudle
<fjh> please indicate the dates so that I can work with XML Security
WG to publish Signature Properties
AB: we never agreed on specific dates but talked about LCWD in April
and CR in June
RESOLUTION: the Widgets Digital Signature spec is Feature Complete;
we will not add any new functionality
<fjh> so I will ask XML Security WG to agre to publish Signature
Properties next week in XML Security WG call
<arve> we'll dial in anew
AB: I will start a new discussion on when to publish the LC of
DigSig
<arve> we can't seem to dial in again
AB: that's it for DigSig for today. Thanks again FH for your great
work here!
<mpriestl> + 1 on the thanks to fjh
<fjh> thanks, thanks Mark, Marcos
<fjh> as well
P&C: Dropping screenshot
AB: Marcos proposed
([22]http://lists.w3.org/Archives/Public/public-webapps/2009AprJun/0
197.html) dropping screenshot for v1 and has already added it to the
V2 feature list
([23]http://www.w3.org/2008/webapps/wiki/Widgets2_UC%26R). My
recollection is all comments on this proposal were positive. Does
anyone object to this proposal?
[22] http://lists.w3.org/Archives/Public/public-webapps/
2009AprJun/0197.html)
[23] http://www.w3.org/2008/webapps/wiki/Widgets2_UC%26R).
<MikeSmith> ArtB: try again now
AB: any objections to dropping screenshot?
RB: no
RESOLUTION: screenshot will not be in v1 (already added to the v2
feature list)
<MikeSmith> Marcos: is it a local phone problem? or a problem with
Zakim?
<MikeSmith> arve: trying calling back in now, if you can
P&C: <access> element
AB: Robin made a short proposal several weeks ago
([24]http://lists.w3.org/Archives/Public/public-webapps/2009JanMar/0
943.html). There has been a little follow-up on the mail list but
certainly no consensus on a solution. This is one of the major open
issues that is blocking the publication of a new LCWD.
[24] http://lists.w3.org/Archives/Public/public-webapps/
2009JanMar/0943.html).
<MikeSmith> is P11 arve and Marcos ?
AB: Robin, where are we on this?
RB: I think we do have consensus, the issue is on the wording
... I will create some tighter wording
... but feature wise, the comments have been positive i.e. at the
right level
AB: has your proposal been added to the ED?
RB: yes
AB: what's the next step?
RB: there really is no next step
... just need to fix the wording
AB: any comments?
MP: I support the current proposal
... there were some comments from BONDI
<mpriestl> * The User Agent's security policy MAY prevent network
access by the Widget to an IRI that does belong to the set of target
IRIs.
MP: my understanding of the current proposal is the above should be
a MUST
... rather than the MAY
RB: I'm not sure I understand
MP: I'll need to go back and re-read it
AS: the sec policy may be more restrictive about IRI access then the
P+C's access element
MP: yes, I agree with that
... and hence the BONDI statement is correct
AB: Arve, Marcos - do you have any comments about the access
element?
MC: no, we still need to review it
... it appears to be a bit thin; may not cover all of the UCs we
have in mind
AB: please, everyone, send all comments re <access> ASAP!
... anything else on this topic?
... what's the status of Thomas on this proposal?
RB: he may have some issues; not clear yet
P&C: Localization proposal from Marcos
AB: Marcos submitted a comprehensive localization proposal
([25]http://dev.w3.org/cvsweb/2006/waf/widgets/i18n.html) several
weeks ago. The amount of feedback has been very low yet this is a
major open issue that is blocking the publication of a new LCWD.
Currently, only me, Jere and Marcos have expressed their opinions on
the various proposals thus I'd like to hear from other people.
... what do people think about this proposal from Marcos?
[25] http://dev.w3.org/cvsweb/2006/waf/widgets/i18n.html)
[ Silence ]
<mpriestl> (sorry I have to leave the call)
AB: one interpretation of silence is agreement with Marcos
AS: I haven't reviewed it yet
AB: we need feedback ASAP
AS: yes, I understand the priority
AB: what's the next step?
MC: I think we need more feedback
<scribe> ACTION: barstow seek comments from WG members on Marcos'
L10N proposal [recorded in
[26]http://www.w3.org/2009/04/23-wam-minutes.html#action02]
<trackbot> Created ACTION-336 - Seek comments from WG members on
Marcos' L10N proposal [on Arthur Barstow - due 2009-04-30].
AB: we can spend a lot of time on this today
MC: that would be good
AB: Marcos, lead us thru A1 and A2
MC: A1 has no support for sub-lang tags i.e. it does not support
BCP47 and its lookup mechanism
... A2 supports BCP47 lookup
... A2 will reduce the amount of content that must be localized,
particulary if there are multiple levels of sublangs
AB: so there is an efficiency tradeoff here, right?
MC: yes
... Jere and Josh both proposed A2 model
... note that A2 is already in the P+C spec
AB: any comments on A1 versus A2?
MC: I prefer A2
AB: I think we should give people one more week i.e. until April 30
to provide input
... During the Apr 30 call we will decide on each proposal
MC: what if there is disagreement on the 30th
AB: a decision will be made on all of these by the 30th, if not
earlier
MC: David, can we get any feedback from BONDI on the L10N model?
DR: there hasn't been any discussion to this yet
... when do you need feedback?
AB: April 30 is the deadline
... Marcos, can you provide a short description of B1 and B2?
MC: B1 proposes the UA's locale can be a list of locales
... B2 proposes the UA just have a single locale
... HTTP supports multiple languages
... The basic question is: does the UA support one lang or a list of
languages?
AB: any comments on B1 vs B2?
AS: so one use case is about knowing which langs a UA supports?
MC: yes
AS: I would expect the UA to have a single locale but a widget could
be localized in many diff languages
... could then use prefs to manage this
... I can understand a UA support multiple locales but I'm not sure
this is the best way to go about it
AB: yesterday I suggested we need some more reqs work and some more
UC's to understand why need this stuff
[27]http://lists.w3.org/Archives/Public/public-webapps/2009AprJun/02
69.html
... will you reply to my comments?
[27] http://lists.w3.org/Archives/Public/public-webapps/
2009AprJun/0269.html
MC: I think you have some good comments
... we need to get this done as soon as possible
AB: yes, agree. But I also heard Andrew asking for some UCs too
AS: yes; some UCs and Reqs would be helpful in understanding the
problems we are trying to solve
MC: some of the proposals do go beyond what current engines do today
AB: I get concerned about the complexity here for v1
... we seem to be moving from codifying the existing cow paths to
building a new super highway
... Marcos, would you please give us a summary of the C* proposals?
MC: these about deriving te widget's locale
... diff between C1 and C2 is the order of searching for the
widget's locale
AB: I hope that is clear to everyone; any questions?
... what about the D* proposals?
MC: these three proposals are about how to represent the widget's
locale
... my preference is D2
AB: any comments or questions on these 3 D proposals?
[ Discussion between Andrew and Marcos about D2 and various
scenarios ... ]
AB: can we get a short intro to E, F and G proposals?
MC: E proposals are about XML Base and whether we use XML Base
itself or our own emulation of it
AB: I agree E1 seems reasonable to me
AS: does that apply to all URIs?
MC: yes
AS: OK; that's good
MC: F proposal addresses the "missing content" problem
... F1 uses root directory in search; F2 does not use root
... My preference is F1
AB: yes, F1 seems reasonable to me
... I think it works with the principle of least surprise
... please introduce G1 and G2 Marcos
MC: this is similar to F proposals but if the lookup is using URIs
Window Modes spec: status and plans
AB: Robin, what is your level of interest here?
RB: unless someone wants to lead this, I will start working on it
next week
MC: we are interested and willing to collaborate on anyone that
wants to lead it
AB: I think the plan is for Robin to start working on this spec next
week and for MC to help
... is that right?
MC: yes
RB: yes
AB: meeting adjourned
Summary of Action Items
[NEW] ACTION: barstow seek comments from WG members on Marcos' L10N
proposal [recorded in
[28]http://www.w3.org/2009/04/23-wam-minutes.html#action02]
[NEW] ACTION: Barstow work with Frederick re synching Signature
Properties spec with Widgets DigSig spec [recorded in
[29]http://www.w3.org/2009/04/23-wam-minutes.html#action01]
[End of minutes]
