See also: IRC log
NM: http://www.w3.org/2001/tag/2011/02/08-agenda
... Action item review is just checking that we've got the right
things on the schedule in the near term
... Open issue review is quite different, intended to check that we
haven't let things fall between the cracks, or that we are carrying
things we don't need to
NM: http://www.w3.org/2001/tag/2011/02/08-agenda#priorities
... Good for us to review each year where our effort is going, and
how we are going to get it done
... and be sure we have a shared notion of our priorities
... I'd like to get more than one person on the hook for at least
some tasks, to share the work back and forth in some way
... Looking back, we set ourselves some priorities:
Tracking/influencing the HTML work -- hard situation, but we did a
number of things here and I think we did what we set out to do
... We also committed to a Web App Arch effort, since two years,
but I don't feel that we've made as much progress here as I'd hoped
-- we need to look hard at this to see whether we should modify or
even drop our goal
... Third goal was Metadata, an umbrella for many SemWeb issues
JR, LM: No, Metadata is much narrower than that, it is about documents only
TBL: +1 to keeping Metadata narrowly focused
NM: We've also done good work, largely due to LM's efforts, on a number of core web infrastructure issues, including IRIs and media types
LM: I'm actually concerned how little progress on IRIs lately
NM: On the organizational front, we're trying to structure the management of our work via Tracker Products augmented by TAG-specific product pages such as http://www.w3.org/2001/tag/products/apiminimization.html
NM: Tracker has Issues, Actions and
Products
... Actions can be associated with Issues or Products
... See the Guide to TAG Process
NM: Please note that there are two 'Product' pages, one under 2001/tag/products and one under Tracker
NM: Tracker is just not flexible enough to be able to capture in a structured way the information we need at a glance for each effort, and it's also limited in its ability to, e.g., associate an action with an issues and a product.
[Discussion about mechanism, not minuted]
NM: Intent is to have a small number of Products
NM: Need properties for a product: Goals, success criteria, deliverables with dates, schedules, TAG members assigned, related issues.
<timbl> We could do it in RDF if we had a RDF export from Tracker of course
NM: API Minimization is our first
example: http://www.w3.org/2001/tag/products/apiminimization.html
... Goals and Success criteria are the core of these
... Made concrete by deliverables
... Example: ACTION-514: Draft finding on API minimization
LM: I think maybe we need two
categories of Products
... 1) Specific documents or other outputs;
... 2) Things which are more like some of our Issues, e.g. Track
the HTML work
NM: Yes, but can we just try your case (1) for now
TBL: Mechanisms are your business as
chair, the focus is on the content, that's where our energy should
go
... But, having said that, my inner hacker has already built an
ontology for issue/product/... management for the Tabulator
... I could do more hacking and give you everything you want
... In practice lets go ahead as you propose
... But in the background, maybe you and I should try to do
something better
Tutti: Crack on
NM: Regardless of mechanism, do we agree to focus our effort management on setting goals and success criteria, with dated deliverables
<jar> It would be nice if (1) product name could be changed (2) products can be classified somehow (active, complete, etc) (3) notes could be added to product pages
LM: We do other things -- coordination with the IETF
<masinter> want to track the larger theme of W3C/IETF coordination at architectural level
LM: This is a larger theme
NM: For me that's an Issue, about how to coordinate with other bodies
LM: It's not a management issue, it's
a technical issue -- what is the relationship of Web Arch to
Internet Arch
... What's critical for a Product is Success criteria
... And I think we can identify and evaluate progress for this
effort, so it can be a Product
TBL: Wrt Success criteria, include documentation of important properties of the system which need to be preserved
NM: Other things can have ways to identify and evaluate progress, I want to keep Products for things with deliverables
<timbl> http://www.w3.org/2005/01/wf/flow#Task <-- the high-level concept of task
DKA: With respect to TAG priorities, there's also
the W3C 2011 Priorities and Milestones document
... http://www.w3.org/2011/01/w3c2011#Summary
NM: This reminds me that there are two ways to come at our planning: internally-driven and externally-driven
DKA: In particular, are we missing anything from Jeff Jaffe's list?
NM: So take a tentative pass at what
we are already spending time on
... and then see if there's anything we're missing
... at which point we will know if we're over-committed
LM: It's great to see a W3C priority
list of technical topics
... I'd like to respond to it
... So this is higher priority for me than reviewing our current /
past efforts
HST: The chair is asking for help in getting to that, by first clarifying the status of our existing commitments
NM: Here's another Product: HTML/XML Unification Draft Product Page
<masinter> I think the "big theme" here is: architectural coherence of the W3C protocol and format work. And that XML / HTML is a lead element, because so much of W3C work is based on XML and yet HTML consistency with it is at issue and that the TAG could look at whatever the "task force" produces in this context. The goal should not be "Unification" but "coherence" and "support for workflows and use cases" and there are various sub-products, around IRIs and URI schemes....
<noah> ACTION: Noah to build Tracker product page for HTML/XML Unification [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action01]
<trackbot> Created ACTION-522 - Build Tracker product page for HTML/XML Unification [on Noah Mendelsohn - due 2011-02-17].
LM: The big theme here is
architectural coherence between W3C RECs
... I wouldn't want to track this as Unification, because that's
not the goal even for XML vs. HTML
... I don't think that goal stands up
NM: I hear you as observing that
there's a higher theme that this specific Product fits into
... and I think we can do that, we can have Themes
... The name comes from the history -- is the key point the
abstraction of a higher level
LM: Either this fits in one of the
high-level things the JJ laid out, or something else
... in this case, something else, which is a particular TAG
responsibility
NM: I hear this, and will try to find a way to organize our thinking at this level
LM: Pass for now
HST: [proposed minor agenda restructuring]
ISSUE-60: Web Application State Management
AM: speaks to http://www.w3.org/2001/tag/2011/02/ClientSideStorage.pdf
... I need guidance on how to take this forward
<masinter> This underlying architectural issue relates to "Powerful Web Apps", "Data and Service Integration" and "Web of Trust": web applications are more powerful if different applications can share. But they have to do it in a secure way that also maintains user privacy.
AM: The fundamental issue is how to
manage the inevitable intrusion of the Privacy/Security issue into
any discussion of client-side storage:
... 1) Ignore it, and just do the storage thing;
... 2) Try to do the integration.
AM: The answer is different depending on whether we see the deliverable here as stand-alone, or as part of a larger document where Security is being taken care of
TBL: The document talks mostly about cookies, but there are a large number of new technologies, e.g. sqllib, which are at least as important going forward
<masinter> Security sections could move to https://wiki.tools.ietf.org/wg/websec/charters based on http://w2spconf.com/2010/papers/p11.pdf
TBL: And as you talk about privacy in that context, it becomes a question about what 'agent' (software, site, person) can get access to what
AM: You're going beyond data
TBL: No, just data raises these
issues, say I have an RDF store on my phone, and an app written by
an airline is running in a container from a third party and wants
access to that data. . .
... At worst we end up all having to have our own copies of all the
privacy-implicated software, to ensure our data doesn't get
away
TBL: So this discussion has to be forward-looking to address not just what's here now, but what's coming soon
<masinter> "In 2011, W3C expects to charter a Web Application Security Working Group for work on specific technologies to enable more robust and secure Web Applications." from http://www.w3.org/2011/01/w3c2011 under "Privacy and Security"
JAR: Normal engineering practice
should be followed, to look first at the requirements, without
jumping too soon to the technology (e.g. cookies)
... You started out with "need....", which are requirements, and
then jump to security -- but that's a requirement too
... It's like building a LISP interpreter, if you leave memory
management to the end, you end up with a buggy implementation
AM: Right, so you're saying add security as a requirement, early
JAR: Only then do you look at
solutions
... and try to match requirements to aspects of solutions
LM: There is a commitment at W3C level to charter a Privacy and Security WG
<noah> Actually, the slide just said privacy, and I think that's what I heard him ask about. That's why I got confused when we kept talking about security.
LM: And that group is a candidate recipient for this work
AM: I thought it was a Privacy IG
that was on the way
... and that's not quite the same
LM: W3C has committed to chartering a
Web Applications Security WG
... In JJ's document
<noah> From: http://www.w3.org/2011/01/w3c2011
<noah> In 2011, W3C expects to charter a Web Application Security Working Group for work on specific technologies to enable more robust and secure Web Applications.
<noah> (public document)
AM: So, yes, when that happens, feeding in to it makes sense
NM: On the separate vs. together
point (storage vs. Privacy&Security)
... indeed per JAR sometimes it's dangerous to factor
... but not sure that's true here
... Suppose you did just focus on storage, w/o talking about
P&S
<masinter> "Client side state" doesn't really have anything to say unless there is some 'memory' or 'communication' of client side state
NM: What would the Product page look
like if you did that (thought experiment)?
... If you can't even do that, we've learned something
... And if you can, then we can look at the P&S factoring
question as such
... Thinking about the Product page should be really helpful
AM: I want to come back to the "one large document" question
JAR: That's not what I said. . .
NM: If we want to do a large
document, it's a long way out
... So even if we are aiming for a merged form, the work has to go
ahead as if it were going to stand on its own
LM: Different perspective -- we're not designing an implementation -- there are already a number of design patterns for Client-side state, and they differ
LM: they have different relevant
properties to the requirements
... Here are seven different design patterns; here are their
properties, here's why some address requirement X, Y, Z better/worse than
others
<masinter> "seven" plus or minus four
NM: Assuming this is a separate document, what are the top three questions it will answer for the community?
AM: Give me three weeks
NM: OK, let's suspend judgement on the long-term future of this work until we see your response
<masinter> are there books or papers on web application design, that cover client side storage, use of cookies, local storage, etc?
AM: We asked the WebApps guys who are
writing these specs, where are your use cases?
... And they didn't have much of a concrete reply
<noah> ACTION: Ashok (with help from Noah) build good product page for client storage finding, identifying top questions to be answered on client side storage Due: 2011-03-01 [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action02]
<trackbot> Created ACTION-523 - (with help from Noah) build good product page for client storage finding, identifying top questions to be answered on client side storage Due: 2011-03-01 [on Ashok Malhotra - due 2011-02-17].
[Break until 1045]
[resume from break]
NM: I've been reviewing the open
actions, to try to abstract what the set of Products are in
principle
... So that we can create the ones that are missing
... Quick scan of the Tracker Products:
2001/tag/group/track/products
... Agreed that we are not currently working on the Versioning
Product
<noah> ACTION: Noah close versioning product [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action03]
<trackbot> Created ACTION-524 - Close versioning product [on Noah Mendelsohn - due 2011-02-17].
LM: Some of that work is going forward under other headings, e.g. the mime info work
NM: What is this WebApp Access Control product?
JR: Ask JK
<noah> ACTION: Noah to check with John before closing http://www.w3.org/2001/tag/group/track/products/2 WebApps access control [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action04]
<trackbot> Created ACTION-525 - Check with John before closing http://www.w3.org/2001/tag/group/track/products/2 WebApps access control [on Noah Mendelsohn - due 2011-02-17].
<noah> ACTION: Noah to do first draft product stuff for MIME and related core web mechanisms [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action05]
<trackbot> Created ACTION-526 - Do first draft product stuff for MIME and related core web mechanisms [on Noah Mendelsohn - due 2011-02-17].
NM: We have a total of 45 open actions
LM: I want to push Action 519 to be
even bigger, on the relation of standards to operational
requirements
... Big ISPs come to IETF, not to W3C, so this is important with respect to our
presentation to the IAB
<noah> ACTION: Noah to make sure we make progress on ACTION-519 and ACTION-517 in time to provide input to Prague IETF meeting, talk to be ready by mid-March [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action06]
<trackbot> Created ACTION-527 - Make sure we make progress on ACTION-519 and ACTION-517 in time to provide input to Prague IETF meeting, talk to be ready by mid-March [on Noah Mendelsohn - due 2011-02-17].
NM: Diving in to Action-521, do we
want to press forward with taking Disposition of Names in a
Namespace to REC: 4 not sure, 2 against, 1 to push it to Core, 0 to
do it
... Remind NM to propose next steps and/or discussion on this
... Relieved not to find too many "Oops, we've let this slip"
responses or "Oops, there's a big iceberg under here"
... Open for discussion, let's propose edits to the list of
Products
... Additions or deletions
<Zakim> ht, you wanted to say Products don't exhaust our work
<Zakim> jar, you wanted to take apart 'important'
LM: Change HTML 5 review to Open Web
Platform Architecture
... At the TPAC plenary, the MS rep proposed a number of
HTML5-related arch. issues
... and I've gotten a list from Julian Reschke
<masinter> and from several other people
HST: Is Persistence a Product
NM: Should we be doing that -- think about where this stands?
LM: I don't think it is one of the top priorities aligns with the guidance we're getting
<masinter> I'm looking at http://www.w3.org/2011/01/w3c2011
TBL: We are responsible for long-term issues, which no-one else will worry about
NM: I read JJ's list as a "be sure to cover this", not "and nothing else"
HST: We owe it to the people who raised the persistence question to work on it, and I think addressing why people don't trust 'http:' URIs is a fundamental arch. question.
NM: Goals and success criteria
HT: We have two draft documents in
different stages: 1) my somewhat stale but valuable Dirk and Nadia
design a naming scheme and 2) Jonathan's checklist document
... I think each of those speak to a different community, and
suggest different deliverables directed at different goals.
<masinter> the reason why I'm reluctant to put this is a priority is that I'm afraid i have some real disagreements about the nature of the problem and the directions to address them.
HT: Potential goal #1: address the
architectural origins of the vulnerability of Web names.
... Potential goal #2: identify best practices for the use of Web
names in contexts where some form of persistence is goal.
<scribe> ACTION: Henry to create and get consensus on a product page and tracker product page for persistence of names Due: 2011-03-01 [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action07]
<trackbot> Created ACTION-528 - Create and get consensus on a product page and tracker product page for persistence of names Due: 2011-03-01 [on Henry S. Thompson - due 2011-02-17].
<timbl> due date: 3011-01-01 -- test that the action URI still works
ACTION-528 Due 2011-03-01
<trackbot> ACTION-528 Create and get consensus on a product page and tracker product page for persistence of names Due: 2011-03-01 due date now 2011-03-01
<masinter> "persistence" requires both technical and social institutions to coordinate. We should look at successful social institutions and those in trouble.
<masinter> http://www.archive.org/post/337580/internet-archive-needs-your-help
DKA: Offline web: widgets, app cache, cf. JJ's Web Apps and mobile devices bullet
DKA: There is a workshop being organized by Matt Womer in this area
NM: This overlaps with Client-side state
DKA: This is about packaging
... not (just) storage
NM: Should we discuss making this a
product?
... OK, will do
<noah> ACTION: Noah to schedule telcon discussion of a potential TAG product relating to offline applications and packaged Web [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action08]
<trackbot> Created ACTION-529 - Schedule telcon discussion of a potential TAG product relating to offline applications and packaged Web [on Noah Mendelsohn - due 2011-02-17].
NM: All of mobile?
DKA: No, mobile and the offline web -- packaging the web
<Ashok> Interacts with Client-Side Storage
JAR: Saying something is important is
not very useful, unless someone is signed up for it
... Maybe we should do a gap analysis: a matrix where we have
supply-side -- what would each member be inclined to do, left to
themselves, vs. demand-side: what have JJ and/or our community
asked us to do
... and we look for the blank spaces
... And we don't yet have enough information yet to actually build
that matrix
NM: That's a goal for us, yes
<masinter> alignment between W3C working groups, and with IETF and with previous specs and .... is after all what TAG was originally chartered for
<Zakim> masinter, you wanted to talk about 'underlying architecture' as possibly a higher TAG priority than Jeff's list, which applies to W3C as a whole
<Zakim> timbl, you wanted to wonder about a goal in which social institutions are changed in order to achieve persistence.
<noah> Henry and Larry will be there.
AM: Talk or panel.
LM: See ACTION-500. There is a panel, with representation from lots of the IETF community. Panel description is copied in the action.
LM: Not yet determined between Henry and me who will actually be on the panel.
ACTION-500?
<trackbot> ACTION-500 -- Larry Masinter to coordinate about TAG participation in IETF/IAB panel at March 2011 IETF -- due 2011-02-15 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/500
AM: You probably only get 15 minutes?
LM: At most, could be 10.
... We should use this mainly to "show the flag", indicate where
major points of interest are, etc.
... They've written what they think the issue is for them.
HT: It's in some sense better we
don't have a longer slot, which would lead to us reading our
laundry list.
... The appropriate question we need to think of here today is,
what do we want to project about the TAG itself?
LM: We are in the process of establishing our priorities based on what the community needs from us. Some people at the IETF meeting are likely to be, unfortunately, not W3C members.
NM: Um, our TAG community is the Web and Internet community, not just the W3C.
LM: Oops, you're right, that's what I meant.
NM: We listen to everyone, on www-tag, by inviting people to join meetings and calls, etc.
HT: The IETF is appealingly a
crypto-anarchist commune with a long history.
... They are phenomenally successful.
... Larry and I should probably send email to www-tag asking for
input, then get telcon time.
LM: Henry, hows about you draft a talk for review, with my help?
HT: I'll produce say, 5 slides, for review on call in two weeks.
<masinter> what is the tag, what the tag works on, what things are we thinking about in W3C, what things are we thinking about in the TAG in particular
<scribe> ACTION: Henry to draft slides for IETF meeting, with help from Larry Due 2011-02-22 [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action09]
<trackbot> Created ACTION-530 - Draft slides for IETF meeting, with help from Larry Due 2011-02-22 [on Henry S. Thompson - due 2011-02-17].
NM: Suspended for lunch
Philippe Le Hégaret joins the meeting
Discussion of action items
NM: Larry asked me to add a link to RFC5226 to the agenda.
<noah> http://tools.ietf.org/html/rfc5226
<noah> http://lists.w3.org/Archives/Public/www-tag/2011Feb/0004.html
DQA: I note IE9 has Geolocation.
<masinter> there was another link http://tools.ietf.org/html/draft-iab-extension-recs
Larry:re ACTION-511: Send email framing TAG work on registries
LM: we have had a lot of
discussion of registries
... perhaps as reaction to IANA, feeling that registries were
... a bottleneck in the system, that we should use URIs to be
decentralized.
LM: Still, there are protocols,
protocol and language elements where we don't use URIs.
... But, if it isn't a URI, then how do you find out what it
means?
<plh> --> http://www.w3.org/2005/04/xpointer-schemes/ XPointer Registry
LM: Does IANA still manage it? But
IANA is unresponsive and cumbersome? Should we use a wiki page,
[HTML WG suggestion]?
... I was trying to frame the issue with MIME type registries.
<plh> --> http://www.w3.org/2002/06/registering-mediatype.html Register an Internet Media Type for a W3C Spec
LM: Many issues are around what
the mime type means when it evolves, having to do with
versioning.
... There are technical and social issues. Power: who controls the
registry? Who controls what properties things should have
registered?
... People disagree on the contents of the registry
... I pointed to RFC2434, now RFC5226 .
LM: I also saw a good IANA document in progress on extensibility from the point of view of protocol design, in which registries are one way.
<masinter> http://tools.ietf.org/html/draft-iab-extension-recs
PLH: I pasted in various links,
including to the XPointer registry.
... This registry is hosted by W3C.
HT: The XPointer spec didn't have unqualified names, but people complained that getting URIs in to bind every name was ridiculous. Please let us defined some short names which we can own, and we did, and so we have a URI-based registry mechanism.
<plh> http://www.w3.org/2005/04/xpointer-schemes/range
HT: the way you tell what short names mean or are available is you concatenate with a URI.
PLH: This was very lightweight,
lightweight review process too.
... We demand a link to a spec but no other review.
HT: Just a way of mapping short names into URI space on a first come, first served basis.
LM: What does CSS do with vendor prefixes?
Peter: Nothing formal -- we have recently started keeping a list.
NM: Is it just a convention?
<timbl> ... You register just the -moz- not the -moz-* names.
PL: No, more than that. The spec requires a syntactic convention for use of anything that is either not in the spec, or not advanced to a certain point in the spec development.
TBL: Do you standardize thinks like -*-roundedcorner?
PL: No, just -*-
TBL: As a CSS user, having many different names was a pain for Rounded Corners.
Peter: That was necessary as the different vendors did it differently.
Larry: We were having registries, so
we are not really following out URI architecture. Can IANA be fixed?
Is the problem IANA?
... People say the problem is not IANA but tracking what IANA is up
to.
<plh> --> http://dev.w3.org/html5/status/issue-status.html#ISSUE-027 HTML ISSUE 27
TBL: For example, the text/n3 mime
type is still pending
... after years
Larry: if you look at the docs establishing how IANA works, they don't determine the process ... that is established for each registry anew. I refined the URI scheme registry process, there is still unhappiness with it.
LM: I would hate for W3C to reinvent this wheel and rediscover all the problems
PLH: This is related to infamous HTML
WG Issue 27 (see link above)
... (all HTML WG issues are infamous)
PLH: One proposal is to have a registry at W3C
<masinter> proposal W3C run rel: http://www.w3.org/html/wg/wiki/ChangeProposals/RelRegistryAtTheW3C
PLH: Mark Nottingham has done work on
a IANA registry. Ian Hickson tested it and declared that it was not
working.
... there is a counter-proposal which just uses a wiki page.
... This was escalated to the WG as issue 27.
<masinter> http://tools.ietf.org/html/rfc5988
<noah> http://www.w3.org/html/wg/tracker/issues/27
<noah> ISSUE 27: @rel value ownership, registry consideration
Larry: We should discuss whether and why and how W3C runs registries -- it should not be decided just by a local WG, as it is a long term commitment, and much more than the design of a technical spec.
PLH: Without requirements, you can't
PLH: It took years to get
image/svg+xml took years to get registered.
... Even though it was in use for years.
<plh> --> http://www.ietf.org/mail-archive/web/ietf-announce/current/msg08275.html Approval of image/svg+xml Media Type
Larry: People brought this up as a poster child of why it didn't work ... but they didn't in fact respond to IANA's comments about what was missing from the application
<masinter> there's also been a long recent discussion about +json and +zip; and +xml is an issue
TBL: We had a story with text/n3+rdf
type where we used the W3C/IETF liaison meeting to track. Per that
discussion we removed the +rdf.
... They said we would have to produce a stable document, which we
did some years ago, so for me text/n3 is another poster child for
the problems.
<plh> --> http://www.w3.org/TeamSubmission/n3/ N3
TBL: The confusion is compounded because there are people out there using the now deprecated +rdf form, but there's nothing to point to saying, "here's what you should do".
<masinter> Maybe W3C should have an IANA shepherd who knows how to work IANA and helps people through the process, that would be better than running W3C registry..
<plh> for n3, I'm probably the bottleneck
TBL: There's also no tracker for the
application review process for mime types. You can't tell where
things are in the process, what the problems are, or even that
there is a registration pending.
... So, one suggestion is that we should not only run a registry at
W3C, but that we should run a tracker.
LM: You could run a tracker for
IANA
... The IETF tools team has built tools for many groups, and
perhaps has just not gotten to IANA
LM: The IETF tools team has been building tools for IANA but not that one yet.
PLH: The technical issues we have to
resolve, and they can take years
... The charset attribute, and then content-encoding, the discussions
exhausted the energy of the applicants.
Larry: My experience has been very
positive: you tell the truth you get approval. With text/html Dan
Connolly and I updated it... I also did application/pdf.
... I was involved with gopher's mime types
... What can take years has been miscommunication.
<plh> --> http://www.imc.org/ietf-xml-mime/mail-archive/msg00981.html MIME Type Review Request: image/svg+xml November 2004
TBL: I sympathize with the
requirements they have for, e.g. MIME registry, but I've found that
the HTML experience of having two specs (I.e. the HTML spec plus
the media type registration), was not good. We've now fixed that by
ensuring that the spec shall pass muster as a registration
document, and IANA will please accept that.
... That now is the case, which is good.
... Therefore, my view is that the right path for SVG would have
been that all the stuff like charset should have been caught and
fixed as part of the W3C CR process reviews.
<Zakim> jar, you wanted to mention journals e.g. PLoS One
JAR: This is not happening in a vacuum
-- there have been registries before IANA
... It isn't just who runs it, it is what properties it has:
... What criteria of acceptance, professionalism of management,
what tracking technology,... the publication of a scholarly journal is an
analogous process, [foo???] example.
LM: We use registries for extensibility, where the spec points to a given specific registry, an the standard defined the criteria for the registry, so that the standard will still work. If someone tries to register a term which violated the design, then it is rejected.
<masinter> maybe this is an important criteria for registries -- that the protocol design shouldn't rely on the registrar review to maintain invariants
Tim: Example -- HTTP headers always, per RFC822, have a comma -as an equivalent to a new header line - the cookie header spec in error used it differently and it was not caught.
Larry: The spec puts an onus on the good people running the registry to make sure that good things happen.
LM: In some cases in the past, the spec did not tightly bound what extensions could do, and we relied on the registrar to enforce good practice.
TBL:Hmm. I'm sure Larry is right about the history, but it seems preferable to me that the spec should say what extension points can do, and the the registrar merely enforce that
<masinter> http://www.w3.org/2002/06/registering-mediatype.html Register an Internet Media Type for a W3C Spec
PLH: We have a media type registry at W3C
PLH: Since M Duerst left w3t, I have
been maintaining the big table at the bottom
... This table has been there for 8 years
... The old way of registering a media type is to just write an
RFC, but a few years ago, with Martin's help, IETF allows other
organization's specs to be used in the IANA registration.
<plh> --> http://www.w3.org/2002/06/registering-mediatype.html#RegStatus Status of Internet Media type registrations
TBL: Is N3 in the table?
PHL: No, my fault. Kick me.
TBL: Will do.
PLH: I accept total responsibility
for making sure that it is
... Many of these media types are here but not in the IANA
registry.
Larry: How many of these have been requested?
PLH: If you look at the "Plans" column.
TBL: I suggest that the states be defined in an ontology
PLH: "Need IETF types review" means that W3C has yet to ask for that review.
[discussion of W3C process]
PLH: We have those steps to help
working groups go through those processes.
... We can end up with things which just hang there
HT: What is the problem we are trying to fix now?
PLH: The problem with SVG was getting is registered in 2010 after asking in 2004, with it being used in between.
PLH: For me the problem is that we requested an SVG media type in 2004, that only got formal approval in 2010, and it was used without registration for 6 years.
HT: OK, stipulate a problem with that registry, the TAG issue appears to be about registries in general.
HT: Sounds like a bug in that registry -- lets suggest that they implement a tracker. That could be fixed. Automating the registry wouldn't necessarily help that. The XPointer scheme registry has a rule that the URI works and tells you the status the moment you have requested registration, but that's a management decision, not a technical one.
Larry: It would be nice to give IANA a heads up before the request -- an intent to register. You could post that they intend to register it.
Tim: propose that the IANA system should surface all the info in PLH's table
<masinter> http://www.w3.org/2002/06/registering-mediatype.html
<masinter> but if OASIS and ISO and other organizations want to register values, shouldn't they also be visible to W3C members?
Larry: There is a place for
lightweight registries, e.g. MIME types, that many organizations can contribute
to.
... W3C should try to fix IANA before running around it.
... We should volunteer to help them, and find a good way to
integrate the web architecture of the registry with the Internet
Architecture people.
... With specific technical details, for example there are issues about the
MIME types conflicting with the sniffing documents.
Noah: Do we want any more work on this?
Larry: PLH is on the front line, who is being asked to run registries. As the TAG we can help out with arch issues.
PLH: The immediate issue is issue 27, which is related to rel=""
NM: To clarify, I was asking whether we needed to schedule or track work that's beyond what we're already doing
PLH: The next step is if there are counter-proposals in the HTML WG.
PLH: Potentially, the TAG might have a position to offer to the HTML WG
TBL: I'm not sure I'm hearing anyone around the table complain about anything.
JAR: There are RFCs which point to the IANA registries.
<masinter> http://tools.ietf.org/html/rfc5988
JAR: We don't want two registries.
TBL: Right, not two registries, and we want a good relationship with IANA. We do need something that will produce RDF.
JAR: Um, that can be a tarpit. I've already tried to convince IETF on that.
<jar> well, not on exactly that, but on something closely related having to do with link relations and 200 status
TBL: There are, e.g. ontologies that
list each of the HTML headers. People are producing ontologies that
are 1:1 with the IANA registries. What's crucial is to deal in URIs
that you can dereference to find out what you've got.
... IANA spent a long time working in plain text not HTML, a long
time using ftp vs. http, they've slowly moved. I fear we might be
talking a long time to make the move on conneg that returns
RDF.
<masinter> I think people ascribe to "IANA" things that are really within their own control
<plh> --> http://www.w3.org/html/wg/wiki/ChangeProposals/RelRegistryAtTheW3C#Positive_Effects Effects of a registry at W3C
<masinter> there's no reason why W3C can't run a service for doing something with IANA registered terms, for example, by adding to the registry a set of "registered value retrieval services"
TBL: Meanwhile, there are cases where you want to pick up information etc. about a new media type dynamically, while browsing.
NM: Trust issues aside, you could even dynamically pick up handlers, e.g. to render a new image type.
TBL: Indeed, a very interesting rathole, but not now.
<timbl> The relationship between a MIME type and a typical file extension is important for security -- you must not store a file in a file system so that it looks as though it has a different MIME type, as that is a security hole.
ACTION-511?
<trackbot> ACTION-511 -- Larry Masinter to send email framing TAG work on registries -- due 2011-01-20 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/511
PLH: Henry Sivonen suggests a very lightweight system for rel values, similar to the XPointer registry.
Larry: I think i hear enough technical and architectural issues and I am thinking of writing a finding about it.
<noah> ACTION: Larry to write draft document on architectural good practice relating to registries Due 2011-04-19 [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action10]
<trackbot> Created ACTION-531 - Write draft document on architectural good practice relating to registries Due 2011-04-19 [on Larry Masinter - due 2011-02-17].
----------------------------------------------
NM: What does "open" mean of an
issue?
... For those we are not working on actively , we should
categorize them.
... We should close the ones which have been overtaken by
events.
NM: re Issue-7
<noah> ISSUE-7: (1) GET should be encouraged, not deprecated, in XForms(2) How to handle safe queries (New POST-like method?GET plus a body?)
<trackbot> ISSUE-7 (1) GET should be encouraged, not deprecated, in XForms(2) How to handle safe queries (New POST-like method?GET plus a body?) notes added
Is this still relevant?
Larry PING attributes ping a server to show you took a link
Larry: It might be in the WHATWG spec
still.
... but not in the W3C spec.
... This battle has been fought.
LM: We should worry about the W3C spec.
NM: Disagree, at least in principal. If any organization is promoting widespread use of something we consider inappropriate, that's potentially of concern to the TAG.
TBL: Yes, but we have to pick our battles.
HT: What about the original XForms issue.
HT: Is XForms actually using GET? Many of those who use it use POST not GET, and that is how
XForms architecture is designed to work.
... I didn't realize there is a tension there.
<masinter> I defined MIME type multipart/form-data in http://www.faqs.org/rfcs/rfc2388.html
HT: But XForms uses POST just in order to have an XML body.
Larry: Lets close this without prejudice.
TBL: Let's close it without prejudice
NM: Fine with me
<timbl> TrackBot, Close ISSUE-7
<trackbot> ISSUE-7 (1) GET should be encouraged, not deprecated, in XForms(2) How to handle safe queries (New POST-like method?GET plus a body?) closed
RESOLUTION: We will (re)close ISSUE-7, without prejudice with respect to HTML ping being good/bad
close ISSUE-7
<trackbot> ISSUE-7 (1) GET should be encouraged, not deprecated, in XForms(2) How to handle safe queries (New POST-like method?GET plus a body?) closed
<ht> It appears that @ping has been removed from HTML5[W3C], remains in HTML[WHATWG], but is not receiving much (any?) implementation http://lists.w3.org/Archives/Public/public-html/2010Mar/0136.html
<ht> This is from HTML WG issue 1 http://www.w3.org/html/wg/tracker/issues/1
----------------------------------------------
http://www.w3.org/2001/tag/group/track/issues/20
ISSUE-20: What should specifications say about error handling?
<trackbot> ISSUE-20 What should specifications say about error handling? notes added
HT: If this is being pursued it would be in the XML HTML TF
<noah> Last status change was: connecting with "HTML 5 review" product a la http://www.w3.org/2001/tag/2009/11/02-agenda
HT: Propose this has been overtaken by events.
HT: I think this is overtaken or subsumed with respect to/HTML.
LM: Those are specific instances, but there's a broader concern here.
Larry: Those are specific instances
-- we have though a general question of conservative/liberal, error
handling etc. here.
... Like, if you dictate what happens exactly with every error,
are they still errors?
HT: On a scale of 1..10, that concern
is for me a 2
... in terms of its importance to the TAG.
<noah> http://lists.w3.org/Archives/Public/www-tag/2003Dec/0044
Noah: Look at the history. We closed
in in 2003 - Chris L in 2003 -- the TAG closed it in 2003
... In 2008, on Dec 9, we re-opened it specifically about HTML5 Tag
Soup.
... So HT's comment does indeed carry the day.
Tim: Suggest open, work happening in XML HTML task force.
<masinter> mark it as "PENDING REVIEW"?
<noah> Added note to ISSUE-20: Reviewed status of this at 10 Feb 2011 (8-10 Feb) F2F. Decided to leave this open for now, pending better understanding of where the XML/HTML Unification Task force is going with related issues.
----------
Noah: What about Issue-24
Larry: Lets leave it open
Noah: Issue-25 Deep Linking -- any actions
DKA: I made a very sketchy draft I made -- needs discussion
Noah: Stays open, you have an action for it.
<DKA> http://www.w3.org/2001/tag/doc/rightToLink.html
JAR: Issue-31 was re-opened for UMP.
Noah: Issue-31 stays open. Action-344 now is associated with it
<masinter> issue-31?
<trackbot> ISSUE-31 -- Should metadata (e.g., versioning information) be encoded in URIs? -- open
<trackbot> http://www.w3.org/2001/tag/group/track/issues/31
<DKA> http://www.w3.org/TR/2007/CR-WICD-20070718/
Noah: We close this as no objections heard
<masinter> issue-33?
<trackbot> ISSUE-33 -- Composability for user interface-oriented XML namespaces -- open
<trackbot> http://www.w3.org/2001/tag/group/track/issues/33
<noah> RESOLUTION: Closing ISSUE-33 because CDF is gone, and any concerns about SVG, MathML, etc. in HTML are being tracked elsewhere.
<noah> close ISSUE-33
<trackbot> ISSUE-33 Composability for user interface-oriented XML namespaces closed
------------
<masinter> issue-34?
<trackbot> ISSUE-34 -- XML Transformation and composability (e.g., XSLT,XInclude, Encryption) -- open
<trackbot> http://www.w3.org/2001/tag/group/track/issues/34
Issue-37?
<trackbot> ISSUE-37 -- Definition of abstract components with namespace names and frag ids -- open
<trackbot> http://www.w3.org/2001/tag/group/track/issues/37
<masinter> issue-39?
<trackbot> ISSUE-39 -- Meaning of URIs in RDF documents -- open
<trackbot> http://www.w3.org/2001/tag/group/track/issues/39
<noah> "The community needs:
<noah> A concise statement of the above architectural elements from different specs in one place, written in terms which the ontology community will understand, with pointers to the relevant specifications."
JAR: I wondered about opening an Issue for Harry Halpin's concerns. The problem with doing # or 303.
timbl: Let's not re-define issues under the same number, that's fraud :-)
<noah> ACTION: Jonathan to propose changes to status of issue-39 & issue-57, and perhaps opening new issue relating to H. Halpin's concerns about 200 responses Due: 2011-02-22 [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action11]
<trackbot> Created ACTION-532 - Propose changes to status of issue-39 & issue-57, and perhaps opening new issue relating to H. Halpin's concerns about 200 responses Due: 2011-02-22 [on Jonathan Rees - due 2011-02-17].
<noah> Day 1: Dan
<noah> Day 2: Larry
<noah> Day 3: Henry
[BREAK]
Noah: Now going through action items
<noah> http://www.w3.org/2001/tag/group/track/actions/open?sort=owner
Noah: Now going through action items
Action-505?
<trackbot> ACTION-505 -- Daniel Appelquist to start a document with respect to issue-25 -- due 2011-01-25 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/505
DKA: Do we need a TAG finding here?
Noah: Take us to the point where we are ready for discussion.
DKA: I need someone to help me on this
JAR: We could talk.
<noah> At Feb 2011 F2F, Jonathan agrees to give Dan a bit of help. Next goal is for them to take us to the point where we are ready for telcon discussion.
<noah> ACTION-505 Due 2011-03-01
<trackbot> ACTION-505 Start a document with respect to issue-25 due date now 2011-03-01
Action-507?
<trackbot> ACTION-507 -- Daniel Appelquist to with Noah to suggest next steps for TAG on privacy -- due 2011-03-01 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/507
DKA: We didn't come up with a product page for the over-arching product on privacy.
Noah: The product page is to define work the TAG will do.
action continues.
<noah> ACTION-460 Due 2011-03-08
<trackbot> ACTION-460 Coordinate with IAB regarding next steps on privacy policy due date now 2011-03-08
<noah> ACTION-480 Due 2011-03-01
<trackbot> ACTION-480 Draft overview document framing Web applications as opposed to traditional Web of documents Due: 2010-11-01 due date now 2011-03-01
<noah> ACTION-116?
<trackbot> ACTION-116 -- Tim Berners-Lee to align the tabulator internal vocabulary with the vocabulary in the rules http://esw.w3.org/topic/AwwswDboothsRules, getting changes to either as needed. -- due 2011-02-11 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/116
JAR: Tim took this on himself, up to him whether to proceed
TBL: OK, maybe this is overtaken by events
Agreed on Feb 10 2011 at F2F Jonathan will move this to become an AWWSW action
close ACTION-116
<trackbot> ACTION-116 Align the tabulator internal vocabulary with the vocabulary in the rules http://esw.w3.org/topic/AwwswDboothsRules, getting changes to either as needed. closed
ACTION-510?
<trackbot> ACTION-510 -- Tim Berners-Lee to write a note conveying the TAG's concerns re: the microdata -> RDF URI mappings in the HTML5 microdata draft Due: 2011-01-20 -- due 2011-01-13 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/510
ACTION-510 Due 2011-03-09
<trackbot> ACTION-510 Write a note conveying the TAG's concerns re: the microdata -> RDF URI mappings in the HTML5 microdata draft Due: 2011-01-20 due date now 2011-03-09
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 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/355
ACTION-504?
<trackbot> ACTION-504 -- John Kemp to make sure ACTION-355 links all significant writings including use cases. -- due 2011-01-27 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/504
note that 504 is linked to 355
JK: Unclear whether anyone is interested.
NM: We could do a product page. Could be one with resource assigned and dates, or could be a partial product page, with blanks for assigned resource and dates
JK: Originally, the idea was to fill
out a piece that is called out as missing in AWWW, I.e. to cover
non-HTTP interactions.
... I think that was Noah's original suggestion
JAR: At least, let's not let this get lost
<timbl> http://code.google.com/web/ajaxcrawling/docs/getting-started.html
close ACTION-504
<trackbot> ACTION-504 Make sure ACTION-355 links all significant writings including use cases. closed
ACTION-416?
<trackbot> ACTION-416 -- John Kemp to work on diagrams in "From Server-side to client-side" section of webapps material -- due 2011-03-01 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/416
JK: That's in Ashok's Web App
document. I've made no recent progress.
... What to do whether you will work on future Web applications
document. Ashok now has control of the pertinent document.
NM: Ashok, do you have an action associated with that.
<johnk> http://www.w3.org/2001/tag/2010/05/WebApps.html
<johnk> ACTION-417?
<trackbot> ACTION-417 -- John Kemp to frame section 7, security -- due 2011-01-25 -- CLOSED
<trackbot> http://www.w3.org/2001/tag/group/track/actions/417
http://www.w3.org/2001/tag/group/track/actions/480
close ACTION-416
<trackbot> ACTION-416 Work on diagrams in "From Server-side to client-side" section of webapps material closed
ACTION-508?
<trackbot> ACTION-508 -- Larry Masinter to draft proposed bug report regarding interpretation of fragid in HTML-based AJAX apps Due: 2011-01-03 -- due 2011-02-22 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/508
LM: Discussed Tues.
ACTION-531?
<trackbot> ACTION-531 -- Larry Masinter to write draft document on architectural good practice relating to registries Due 2011-04-19 -- due 2011-02-17 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/531
ACTION-515?
<trackbot> ACTION-515 -- Larry Masinter to (as trackbot proxy for John) who will publish http://www.w3.org/2001/tag/2011/02/security-web.html, slightly cleaned up, with help from Noah and Larry Due: 2011-03-07 -- due 2011-02-15 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/515
ACTION-525?
<trackbot> ACTION-525 -- Noah Mendelsohn to check with John before closing http://www.w3.org/2001/tag/group/track/products/2 WebApps access control -- due 2011-02-17 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/525
ACTION-529?
<trackbot> ACTION-529 -- Noah Mendelsohn to schedule telcon discussion of a potential TAG product relating to offline applications and packaged Web -- due 2011-02-17 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/529
close ACTION-513
<trackbot> ACTION-513 Do F2F agenda closed
ACTION-501?
<trackbot> ACTION-501 -- Noah Mendelsohn to follow up on whether GeoLocation finds reasonable answer on giving permission per site/app etc [self-assigned] -- due 2011-03-01 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/501
ACTION-379?
<trackbot> ACTION-379 -- Noah Mendelsohn to check whether HTML language reference has been published -- due 2011-02-08 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/379
ACTION-379 Due 2011-03-09
<trackbot> ACTION-379 Check whether HTML language reference has been published due date now 2011-03-09
<masinter> why isn't this document listed in http://www.w3.org/html/wg/
ACTION-344?
<trackbot> ACTION-344 -- Jonathan Rees to alert TAG chair when CORS and/or UMP goes to LC to trigger security review -- due 2011-02-15 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/344
Leave for now, moving ahead.
ACTION-532?
<trackbot> ACTION-532 -- Jonathan Rees to propose changes to status of issue-39 & issue-57, and perhaps opening new issue relating to H. Halpin's concerns about 200 responses Due: 2011-02-22 -- due 2011-02-17 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/532
ACTION-381?
<trackbot> ACTION-381 -- Jonathan Rees to spend 2 hours helping Ian with http://www.w3.org/standards/webarch/ -- due 2011-02-11 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/381
ACTION-509?
<trackbot> ACTION-509 -- Jonathan Rees to communicate with RDFa WG regarding documenting the fragid / media type issue -- due 2011-01-29 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/509
JAR: I've been working with Manu Sporny
ACTION-509 Due 2011-03-15
<trackbot> ACTION-509 Communicate with RDFa WG regarding documenting the fragid / media type issue due date now 2011-03-15
ACTION-509 Due 2011-02-15
<trackbot> ACTION-509 Communicate with RDFa WG regarding documenting the fragid / media type issue due date now 2011-02-15
ACTION-477?
<trackbot> ACTION-477 -- Henry S. Thompson to organize meeting on persistence of domains -- due 2011-03-15 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/477
ACTION-33?
<trackbot> ACTION-33 -- Henry S. Thompson to revise naming challenges story in response to Dec 2008 F2F discussion -- due 2011-01-31 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/33
ACTION-33 Due 2011-03-08
<trackbot> ACTION-33 revise naming challenges story in response to Dec 2008 F2F discussion due date now 2011-03-08
ACTION-440?
<trackbot> ACTION-440 -- Henry S. Thompson to ask Hixie what is meant in this [section 9.2] by "retrieving an external entity" and could some clarification be added. -- due 2011-02-01 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/440
ACTION-440 Due 2011-02-22
<trackbot> ACTION-440 Ask Hixie what is meant in this [section 9.2] by "retrieving an external entity" and could some clarification be added. due date now 2011-02-22
ACTION-23?
<trackbot> ACTION-23 -- Henry S. Thompson to track progress of #int bug 1974 in the XML Schema namespace document in the XML Schema WG -- due 2011-01-19 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/23
HT: Reviewed state of this, saw
something on the XML Schema mailing list implying done, but found
closed in error.
... The bit we care about still hasn't been, I'm still
monitoring.
ACTION-23 Due 2011-05-01
<trackbot> ACTION-23 track progress of #int bug 1974 in the XML Schema namespace document in the XML Schema WG due date now 2011-05-01
ACTION-421?
<trackbot> ACTION-421 -- Henry S. Thompson to frame the discussion of EXI deployment at a future meeting -- due 2011-01-21 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/421
HT: I was asked to find out the deal
on deployment.
... Sent a note to the list and got an answer from John Schneider.
Please schedule discussion.
ACTION-511?
<trackbot> ACTION-511 -- Larry Masinter to send email framing TAG work on registries -- due 2011-01-20 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/511
LM: I took another ACTION-531, close ACTION-511
close ACTION-511
<trackbot> ACTION-511 Send email framing TAG work on registries closed
ACTION-512?
<trackbot> ACTION-512 -- Noah Mendelsohn to do F2F local arrangements -- due 2011-01-27 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/512
close ACTION-512
<trackbot> ACTION-512 Do F2F local arrangements closed
<scribe> ACTION: Noah to schedule TAG discussion of !# (check with Yves) [self-assigned] [recorded in http://www.w3.org/2001/tag/2011/02/10-minutes.html#action12]
<trackbot> Created ACTION-533 - Schedule TAG discussion of !# (check with Yves) [self-assigned] [on Noah Mendelsohn - due 2011-02-17].
HT: There are 2 implementations linked from the WG home page, 1 proprietary, 1 open source. Three implementations are reported in the implementation report, but not identified.
We are adjourned