- From: Appelquist, Daniel, VF-Group <Daniel.Appelquist@vodafone.com>
- Date: Fri, 4 Mar 2011 09:40:26 +0100
- To: "tag" <www-tag@w3.org>
- Message-ID: <C99655FA.91D7%daniel.appelquist@vodafone.com>
Hi all please find draft minutes from 03-03 call here:
http://www.w3.org/2001/tag/2011/03/03-minutes.html
Also, text version pasted below. Please let me know if there should be any
revisions.
Dan
---
[1]W3C
[1] http://www.w3.org/
- DRAFT -
TAG telcon
03 Mar 2011
[2]Agenda
[2] http://www.w3.org/2001/tag/2011/03/03-agenda.html
See also: [3]IRC log
[3] http://www.w3.org/2011/03/03-tagmem-irc
Attendees
Present
Larry Masinter, Jonathan Rees, Peter Linss, Noah Mendelsohn,
Henry S. Thompson, Yves Lafon, Daniel Appelquist, John Kemp
Regrets
Chair
Noah Mendelsohn
Scribe
Henry S. Thompson
Daniel Appelquist
Contents
* [4]Topics
1. [5]IAB Panel
2. [6]interaction story for web applications
3. [7]303 related issues.
* [8]Summary of Action Items
_________________________________________________________
<johnk> hmmm, I'm having trouble getting into the call...
<ht> scribe: Henry S. Thompson
Peter: Regrets for next week
<johnk> johnk
<Yves> I read the first two days, and thought they were OK.
<jar_> have scanned the f2f minutes (for lines with my own initials
and a bit more)
Noah: f2f minutes read by anyone?
Jonathan: Scanned, but not read in detail
<johnk> FWIW, I read the first day and thought it was OK
Yves: Read first two days carefully, since I wasn't there, they were
fine
<Larry> +1 approve minutes
Noah: RESOLVED: Approve the 8--10 Feb f2f minutes
<noah> PROPOSE: Approve minutes of 8-10 Feb 2011
[9]http://www.w3.org/2001/tag/2011/02/08-agenda
[9] http://www.w3.org/2001/tag/2011/02/08-agenda
<noah> RESOLUTIO: Minutes of 8-10 Feb 2011
[10]http://www.w3.org/2001/tag/2011/02/08-agenda are approved
[10] http://www.w3.org/2001/tag/2011/02/08-agenda
<noah> RESOLUTION: Minutes of 8-10 Feb 2011
[11]http://www.w3.org/2001/tag/2011/02/08-agenda are approved
[11] http://www.w3.org/2001/tag/2011/02/08-agenda
Noah: Some concerns about the initial draft, please try harder
... Minutes of 24 Feb?
<noah> RESOLUTION: Minutes of 24 Feb 2011
[12]http://www.w3.org/2001/tag/2011/02/24-minutes are approved
[12] http://www.w3.org/2001/tag/2011/02/24-minutes
Peter: I reviewed a bit
Noah: John is coming back for this call for his work item.
... Put one thing ahead - IAB panel.
... Also - at f2f Dan suggested we talk about offline web
application packaging.
... Also we should discuss 303 redirections.
IAB Panel
Noah: anything you'd like to spend time on, Henry?
Henry: Not at this time.
Noah: Anyone else?
Larry: Relationship between scalability and registries - I had some
thoughts.
... We had this issue and discussion on role of registries and IANA.
... We had a discussion on MIME types.
... Architectural issue is preference in webarch for using URIs
rather than registered values (DTD style).
<noah> Good point, Larry
<jar_> jar +1 larry saying: Scalability of URI access relates to the
registry question.
<Larry> I was trying to talk about a somewhat vague thought
connecting work on registries to work on scalability
<noah> What I heard was: if you're going to encourage people to use
URIs for things that otherwise would have been in registries, you
tempt them to make accesses to those URIs, and we've seen that as a
source of scalability problems.
<Larry> if the web architecture prefers using URI-assignment rather
than registry allocation by IANA....
<jar_> E.g. putting the registries and schemas in URI space under
urn: instead of http: might somehow help with scalability question.
Yes?
Larry: In so far as this talk at IETF is to start some discussions
on web architecture and internet architecture: we can have topics we
want to talk about even if we don't have answers.
<Zakim> noah, you wanted to say it's only one bit of the scalability
problem.
<jar_> maybe.
<Larry> well, if the URI used was "data:", there wouldn't have been
a scalability issue
Noah: I see the scalability problem as a fundamental issue for the
web. This type of problem is one concern but not the only one that
might arise.
<Larry> early discussions were about unexpected flash crowds, where
some TV commentator says "look up this cool picture at NASA" and
suddenly NASA's web space is cut down
Noah: For example, the home page for nytimes and cnn - these people
aren't surprised about heavy access, but you could imaging lots of
different resources that might have the same scalability issues.
<Larry> [13]http://en.wikipedia.org/wiki/Content-centric_networking
[13] http://en.wikipedia.org/wiki/Content-centric_networking
Noah: ... when you use URIs, there are scalability issues because
people do [dereference] them inappropriately.
Larry: I am also worried about content-centric networking... would
like to understand this better.
<Yves> scalability issue depends also on cache infrastructures in
the network
MN: [it might be premature to discuss it at the IETF meeting]
<Zakim> ht, you wanted to add that if I put in a slide on this, I
should add two lines about the registry<->URI connection in e.g.
XPointer scheme names
Henry: I think it's important to realise that there are a number of
cases in which the boundaries between registries and URIs have been
blurred.
... It's worth mentioning : we do have a very intentional hybrid
system - the xpointer registry - a database backed registry which
results in a URI being served for everything in the registry.
Larry: Can you give an overview for the panel?
Henry: Yes I think so.
Noah: Moving on to John's topic.
ACTION-355?
<trackbot> ACTION-355 -- John Kemp to explore the degree to which
AWWW and associated findings tell the interaction story for Web
Applications -- due 2011-02-02 -- PENDINGREVIEW
<trackbot> [14]http://www.w3.org/2001/tag/group/track/actions/355
[14] http://www.w3.org/2001/tag/group/track/actions/355
interaction story for web applications
Noah: To frame: identification (URIs), interaction (protocols), ...
... when we started to look at extending work on web arch to
application (as opposed to docuemnts) and we started to see
interactions which are not simple request-response, John undertook
this issue to frame the interaction issues for webapps.
<noah> ACTION-355?
<trackbot> ACTION-355 -- John Kemp to explore the degree to which
AWWW and associated findings tell the interaction story for Web
Applications -- due 2011-02-02 -- PENDINGREVIEW
<trackbot> [15]http://www.w3.org/2001/tag/group/track/actions/355
[15] http://www.w3.org/2001/tag/group/track/actions/355
<johnk>
[16]http://lists.w3.org/Archives/Public/www-tag/2010Jun/0034.html
[16] http://lists.w3.org/Archives/Public/www-tag/2010Jun/0034.html
John: I did an investigation of awww. What I found I sent in an
email to the TAG list.
<noah> Email:
[17]http://lists.w3.org/Archives/Public/www-tag/2011Feb/0058.html
[17] http://lists.w3.org/Archives/Public/www-tag/2011Feb/0058.html
John: the way the interaction model is currently described is over
http.
<noah> Links to document:
[18]http://www.w3.org/2001/tag/2010/10/interaction-examples.html
[18] http://www.w3.org/2001/tag/2010/10/interaction-examples.html
John: some of the things I mentioned were client-side manipulation
and generation of URIs...
... what is the relationship between a server-side application and
the client-side javascript that's running and what enables the
client-side script to know it can construct a URI reliably?
... comet, websocket, ajax-based polling: information rendered to
the user is different than what was downloaded initially.
... In the old model, you had to be running a server to expose a
resource on the web; now you have clients that are servers, also
exposing client resources (e.g. gps) exposed as web resources to
another entity.
... multi-party security is an issue - multiple pieces of content
are mashed up to create a running application.
... More recently I wrote some examples.
... one is the use of websockets; another is the use of geo api to
expose the client's location to the document they've downloaded;
another is client-side URI generation.
<noah> I think when we expose something like an accelerometer using
Javascript APIs as opposed to URIs, then it's best not to call that
a "Web resource". What we have are resources that are linkable
through the mechanisms of the Web, others (like the acceleromotere)
available only at the client, and others that are networked with
non-Web protocols.
John: I think it would be useful to use these examples as a
framework to talk about [webapps architecture]
... All of these things are dependent on an eventing based model
associated with javascript and a document object model that runs on
the client - different from http - so different from what is
document in awww.
Noah: Open floor for discussion.
... How deep and how broad is our investigation of webapps going to
be?
... is this close to a TAG finding?
... doesn't really draw conclusions yet.
... do we want to carry forward with work based on this?
... to elaborate some principles / best practices - terminology for
the abstractions and good practices.
DKA+1 to us building on John's work.
Larry: WebApps are [where it's at]
<jar_> mnot: "Open Source is taking the place of Open Standards"
Noah: Do we have one or two individuals who can work aggressively on
this - 5 to 10 hours a week to write and gain consensus - on this
topic?
Larry: We have a motivation to work on this in terms of starting
some conversations ... at the IETF panel ... IETF has raised some
issues on webapps ...
<jar_> noah would prefer to talk about who is doing the work, rather
than the work.
Noah: We set ourselves a goal of writing a new section of webarch -
new story about interaction. If we're going to write something we
need to write it.
Dan: I think we need to engage with a webapps community of practice
to work on this - worried about being able to do this.
Noah: we should be challenging that community by asking some
questions [ / making some assertions].
... Webarch has good stuff like cool URIs don't change, etc..
provides real advice.
... we should get to that point. Where we can say : here's good
practice and here's bad - and here's useful terminology...
... We should say something specific.
Larry: In the general problem - where we have something to say
that's important but we don't have the resources - could we e.g. ask
the webapps working group what should happen to awww to make it more
relevant to them?
Noah: Goal here is to update the TAG document.
... I'm frustrated we can't find the time to do this.
Larry: What if we publish this as a blog post, ask for suggestions
from the community?
<johnk> I would not want to publish what I've already done as a blog
post
Noah: Chapters suggest terminology, they have principles, good
practices notes...
<jar_> What problem does web architecture solve? ... the answer
would tell us what to do in the apps space.
<Larry> maybe we will get some feedback from IETF meeting on what we
need to do?
Dan: I am happy to reach out the webapps chairs... am worried about
the impactfulness of this proposed document to the community we are
trying to influence.
Noah: We committed to do some work in this space...
... I think you [Dan] are saying the deliverable might be premature.
... then I think we should stop telling the community we're going to
do comprehensive work on webapps.
<jar_> Every journey begins with a single step.
Noah: I am willing to back off on the notion that one of our big
deliverables is a comprehensive webapps architecture.
Jonathan: I think the goal has been a good one -- in that we have
looked at topics [in this space].
Noah: if what we're doing is chaining from "major document" to
"umbrella theme which is influencing a number of point pieces of
work" then we should [be clear on that
... ]
<jar_> Has to do with the TAG status report, setting W3C mgmt
expectations.
Jonathan: There's no crisis here -
... the people who did AWWW felt like there was a real reason to do
it.
... one thing we need here - we should try to figure out what are
the dangers - what are the bad things that might go wrong if we
don't publish this.
Noah: My perception on webarch - the TAG has principles in its
charter; one of these is to document principles of web architecture.
Web apps architecture f[follows on from this]. When you read webarch
and then look at [web apps] [they don't fit together.]
... We should document the web architecture as used today.
Jonathan: I think it's not just a matter of responsibility and
charter - bad things can actually happen and we care about them.
Henry: I don't want to lose this task. If I have time between now
and the end of my time on the TAG this will be the next thing up
because I think it's hugely important.
[discussion on priorities]
Noah: Propose we close ACTION-355 with thanks to John - then see
what else we can propose in the short term.
<ht> ACTION: Noah to work with HST to identify a way forward wrt
interaction [recorded in
[19]http://www.w3.org/2011/03/03-tagmem-minutes.html#action01]
[19] http://www.w3.org/2011/03/03-tagmem-minutes.html#action01
<trackbot> Created ACTION-536 - Work with HST to identify a way
forward wrt interaction [on Noah Mendelsohn - due 2011-03-10].
<noah> ACTION-355?
<trackbot> ACTION-355 -- John Kemp to explore the degree to which
AWWW and associated findings tell the interaction story for Web
Applications -- due 2011-02-02 -- PENDINGREVIEW
<trackbot> [20]http://www.w3.org/2001/tag/group/track/actions/355
[20] http://www.w3.org/2001/tag/group/track/actions/355
close ACTION-355
<trackbot> ACTION-355 Explore the degree to which AWWW and
associated findings tell the interaction story for Web Applications
closed
<ht> action-536 due 2011-08-01
<trackbot> ACTION-536 Work with HST to identify a way forward wrt
interaction due date now 2011-08-01
Noah: Now - proposals on short-term work?
John: Larry mentioned mark N's comments - related to this issue.
... we could link these together....
<noah> ACTION Dan to reach out to Web apps chair to solicit help on
framing architecture (incluing terminology, good practice) relating
to interaction
<trackbot> Created ACTION-537 - Reach out to Web apps chair to
solicit help on framing architecture (incluing terminology, good
practice) relating to interaction [on Daniel Appelquist - due
2011-03-10].
<Larry> hmmm, s/web apps chair/web apps working group/
<jar_> larry email was sent feb 18...
Noah: Anything else under this interaction topic? If not, let's move
on...
<Yves>
[21]http://lists.w3.org/Archives/Public/www-tag/2011Feb/0128.html
[21] http://lists.w3.org/Archives/Public/www-tag/2011Feb/0128.html
Noah: Please put links to this in ACTION-355 and ACTION-356.
<johnk>
[22]http://lists.w3.org/Archives/Public/www-tag/2011Feb/0128.html
[22] http://lists.w3.org/Archives/Public/www-tag/2011Feb/0128.html
<ht> +1 to JR's proposal to regroup under a renamed ISSUE-57
303 related issues.
<jar_> I proposed
[23]http://www.w3.org/mid/AANLkTik5oUpZLs6MVQ5QZEtjqVjLUDWWAo0yLFiXR
9e0%2540mail.gmail.com
[23]
http://www.w3.org/mid/AANLkTik5oUpZLs6MVQ5QZEtjqVjLUDWWAo0yLFiXR9e0%2540mail
.gmail.com
<ht> +1 to JR's proposed new name for ISSUE-57 -- close enough for
government work
Jonathan: I did a survey of URI meaning issues... Rather than
opening a new issue it might be better to use ISSUE-57.
... if we just fix the title and amend it then it will serve
perfectly well.
... I found one caution from Tim.
<jar_> @f2f timbl: Let's not re-define issues under the same number,
that's fraud :-)
Jonathan: but this isn't a redefinition - just a re-titling.
Noah: Do you want to make a case for the scope / new title.
Jonathan: The issue was opened up because of an email to the TAG
regarding 303's - that they weren't working and urging the TAG to
look at other ways to do the same thing.
<noah> [24]http://www.w3.org/2001/tag/2007/07/16-minutes#item06
[24] http://www.w3.org/2001/tag/2007/07/16-minutes#item06
<noah> [25]http://www.w3.org/2001/tag/issues#httpRange-14
[25] http://www.w3.org/2001/tag/issues#httpRange-14
<noah> [26]http://lists.w3.org/Archives/Public/www-tag/2002Mar/0273
[26] http://lists.w3.org/Archives/Public/www-tag/2002Mar/0273
<noah> [27]http://lists.w3.org/Archives/Public/www-tag/2002Mar/0092
[27] http://lists.w3.org/Archives/Public/www-tag/2002Mar/0092
[some discussion on history of the issue]
<noah> At their meeting in 16th July 2007 [$1\47] the TAG resolved
to create a new issue, HttpRedirections-57 as a response to a
community request
<noah> [$1\47]
[28]http://www.w3.org/2001/tag/2007/07/16-minutes#item06
[28] http://www.w3.org/2001/tag/2007/07/16-minutes#item06
<noah> [29]http://www.w3.org/2001/tag/group/track/issues/57
[29] http://www.w3.org/2001/tag/group/track/issues/57
<jar_> [30]http://lists.w3.org/Archives/Public/www-tag/2007Jul/0034
[30] http://lists.w3.org/Archives/Public/www-tag/2007Jul/0034
<jar_> that's giovanni's email which i consider the heart of
issue-57
<Larry> I don't understand what we're talking about and why we're
taking meeting time to talk about it
<Larry> maybe JAR and Noah can take this offline and come back with
one or two proposals for what to do?
Jonathan: the way I think of this - issue-14 was closed with a
decision about how 200s are used - our alternative for those
troubled by this is 303.
... years passed by ...
... then people started saying the solution (using 303) doesn't
work.
... that's a problem that never got fixed - that I'm trying to fix
this year.
... hence issue-57.
<noah> From issue-57 description:
<noah> At the TAG F2F of 4 March 2009
([31]http://www.w3.org/2001/tag/2009/03/04-minutes#item03), the TAG
agreed to "split Issue-57 into two issues as edited by NM, with one
abstention DanC". Issue 62 was opened immediately. Later issue 63
was opened.
[31] http://www.w3.org/2001/tag/2009/03/04-minutes#item03)
<Larry> If people who are trying to deploy something don't like the
implementation consequences of a TAG finding.... it just shows to me
the risk of the TAG coming out with "findings" that propose
technology solutions, without the 'direct' participation of the
implementation community
[discussion on whether or not issue-57 was superseded]
<Larry> and this should be a topic of a working group, not the TAG
<Larry> I have no problem with JAR changing issues to match his
understanding of the issue
<jar_> larry: The TAG made a recommendation (little R) for 303, and
it didn't get review.
<noah> I disagree...it didn't get formal AC review, but it got a ton
of community review (if not complete consensus)
<jar_> larry: People said, we tried it and it didn't work for us...
therefore need a WG
Larry: What should happen now is to tell people who are trying to
engineer solutions : you should form a working group. Because we
suggested a direction, but if it's not working then I don't think
the response should be we should go back and review them. The
response should be : Ok - the thing we recommended has performance
requirements, go and form a working group to come up with something
different.
Noah: It could also be one of the existing semantic web working
groups...
... the community has chosen not to invest before...
Jonathan: Tim has said this is a TAG issue, not specific to RDF.
Noah: Jonathan has made a concrete proposal - an update for issue-57
and an agreement to use that issue to track our upcoming work on
this (which may not be very much).
... going back to Jonathan's specific proposal, I am willing to say
"OK."
<DKA>+1 sounds OK to me.
<Larry> whether it's forming another working group or assigning it
to an existing one?
<jar_> . change per proposal given here
[32]http://lists.w3.org/Archives/Public/www-tag/2011Mar/0000.html
[32] http://lists.w3.org/Archives/Public/www-tag/2011Mar/0000.html
<Larry> note that "Community Groups" in W3C are intended to lower
the overhead of forming a working group
<jar_> thanks larry.
<noah> PROPOSAL:
<noah> 1) Chamge issue-57 title to: At the TAG F2F of 4 March 2009
([33]http://www.w3.org/2001/tag/2009/03/04-minutes#item03), the TAG
agreed to "split Issue-57 into two issues as edited by NM, with one
abstention DanC". Issue 62 was opened immediately. Later issue 63
was opened.
[33] http://www.w3.org/2001/tag/2009/03/04-minutes#item03)
<noah> 2) Add a paragraph to the description per Jonathan's email:
<noah> "On its 2011-dd-dd telcon [$1\47] the TAG noted that members
of
<noah> the community (e.g. in [$1\47]) report that the performance
<noah> characteristics and deployment complexity of using 303
<noah> redirects leave them feeling that they have little option but
<noah> to use 200 responses for this purpose, at variance with the
<noah> TAG's httpRange-14 resolution [$1\47]."
<noah> PROPOSAL:
<noah> 1) Chamge issue-57 title to: "Mechanisms for obtaining
information about the intended
<noah> meaning of a given URI"
<noah> Noodling on this:
Noah: any others worried about use of word "meaning"?
<noah> 1) Chamge issue-57 title to: "Mechanisms for obtaining
information about the referent of a URI"
Larry: You can't ever determine the intended meaning - my worry is
the word "intended."
... A design goal of URIs is to have uniformity of meaning.
<Yves> I am for 'intended meaning', to avoid 'intended semantic'
[debate on the meaning of meaning]
<Larry> i don't like "intended" is that it begs the question of who
intends it
<Larry> depends on what the meaning of 'is' is
<Yves> who intends it... whoever minted the URI
<Larry> issue-57?
<trackbot> ISSUE-57 -- The use of HTTP Redirection -- open
<trackbot> [34]http://www.w3.org/2001/tag/group/track/issues/57
[34] http://www.w3.org/2001/tag/group/track/issues/57
<Larry>
duri:2006:[35]http://www.w3.org/2001/tag/group/track/issues/57
[35] http://www.w3.org/2001/tag/group/track/issues/57
Noah: who prefers meaning and who prefers referent
<noah> referent
<Larry> meaning
<Yves> meaning
<DKA> meaning
<jar_> +1 meaning but not important enought to quibble about
<Larry> actually,
tdb:2006:[36]http://www.w3.org/2001/tag/group/track/issues/57
[36] http://www.w3.org/2001/tag/group/track/issues/57
<jar_> 'individual'
<noah> RESOLUTION: to change tile of issue-57 to Mechanisms for
obtaining information about the intended
<noah> meaning of a given URI
<noah> meaning of a given URI and add para of description per
jonathans email
<noah> RESOLUTION: Change title of ISSUE-57 to "Mechanisms for
obtaining information about the meaning of a given URI" and add
paragrph of description per Jonathan's email
Noah: OK - thanks for your patience with this. Our next call next
week. Let's adjourn for now.
<noah> Jonathan: please leave some tracks in the issue description
to point out when/why it was changed.
Summary of Action Items
[NEW] ACTION 356: [37]Noah to work with HST to identify a way
forward wrt interaction
[37] http://www.w3.org/2001/tag/group/track/actions/536
[NEW] ACTION 357: [38]Dan to reach out to Web apps chair to solicit
help on framing architecture (incluing terminology, good practice)
relating to interaction
[End of minutes]
_________________________________________________________
[38] http://www.w3.org/2001/tag/group/track/actions/537
Minutes formatted by David Booth's [39]scribe.perl version 1.135
([40]CVS log)
$Date: 2011/03/03 23:23:31 $
[39] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm
[40] http://dev.w3.org/cvsweb/2002/scribe/
Attachments
- application/pkcs7-signature attachment: smime.p7s
Received on Friday, 4 March 2011 08:41:11 UTC