See also: IRC log
<trackbot> Date: 02 February 2012
<JeniT> Scribe: Jeni Tennison
<JeniT> ScribeNick: JeniT
plinss: possible regrets next week
noah: Storage discussion will be on agenda next week
Larry: confirm scribe next week
RESOLUTION: Approve minutes of 26 January 2012 http://www.w3.org/2001/tag/2012/01/26-minutes
noah: still missing minutes of 19 January 2012
noah: wanted to check on status of HTML/XML unification report, and on client-side state work, wait for Yves
... also waiting on Dan to provide a section on web apps for report for Jeff
<jar> I urge to expedite
darobin: I can email Dan and see what he has, and try to finalise it
noah: context is Jeff asked for warning about topics that are high-risk
... we went through list at last F2F
... Dan's was on native apps vs web apps on mobile devices
... Yves, we have two notes pending, can you tell us status?
Yves: I've done modifications and am waiting for final OK to request publication on Raman's note
... The HTML/XML report has to get approval, hopefully published Tues
noah: any objections to Yves' draft?
<Yves> http://www.w3.org/2001/tag/2012/01/note/hashinuri.html
RESOLUTION: TAG asks Yves to go ahead and publish note based on Raman's client-side FPWD, and we approve the latest frontmatter from Yves' email
Larry: I'm at the end of what I can do on these actions alone
... they might not be done, but I need guidance from the rest of the TAG about how to take them forward
noah: we have been doing work on mime on the web for quite some time
... first round resulted in Larry publishing IETF draft
... proposal was whether we should do more
<Larry> http://tools.ietf.org/html/draft-masinter-mime-web-info-02
<Larry> action-531?
<trackbot> ACTION-531 -- Larry Masinter to draft document on architectural good practice relating to registries -- due 2011-12-26 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/531
noah: I think Larry wants help here?
Larry: the feedback I got was that maybe there was work on registeries we could do
... from the evolution document
<Larry> http://www.w3.org/2001/tag/2011/12/evolution/Registries.html
JeniT: I thought it was interesting and good work, that I thought we should move forward
<jar> jeni is saying what I was going to say: It's good work, ought to be continued
Larry: this tries to separate out issues about evolution, references and persistence
... we have work on persistence, establishing meaning of URIs, and identifiers that are URIs or identifiers in a registry
... we have some practices around registries that, in the IETF, the registries are managed by IANA
... so there are some documents about IANA registries
... but W3C could sponsor other registries that weren't IANA registries
<jar> managed by IANA *under IETF direction*, right?...
Larry: and maybe there's more to say about that
... I felt like maybe it's a paper for 'Philosophy and the Web' rather than a TAG topic
... I'd like to collaborate with someone else on it
noah: so, there's whether the work shows up as a TAG finding or in a workshop
... and there's whether you do it alone or who's with you
... and any set of combinations makes sense
... so what's your preference?
Larry: I'd like someone else to lead and I'll follow
noah: is there anyone on the call who would like to do it?
ht: "this" is the general status of registries, the goal of registries in web architecture?
<Larry> in particular start with http://www.w3.org/2001/tag/2011/12/evolution/Registries.html
timbl: is it with the same scope as that draft?
<noah> I heard him say "yes, that scope"
Larry: yes, as in that document
JeniT: Pick me, but only after the publishing and linking is done
noah: Publishing & Linking is the one that worries me about winding up needing multiple tries
... I'm worried that if you can't overlap them then this will really put off work on registries
JeniT: I'm interested, but can't take on more work right now
<jar> ditto
<Larry> mainly i want to close this action item
<timbl> I am happy to review it, but can't lead
Larry: let's close the action item, and if someone else takes it on, open another action item
noah: we'll close this action, and I'll take a long-term action that when the copyrighting/linking stuff wraps, we can talk about Jeni taking this on
<noah> close ACTION-531
<trackbot> ACTION-531 Draft document on architectural good practice relating to registries closed
<noah> ACTION: Noah to check, when publishing and linking wraps, whether it's time to reinvest in http://www.w3.org/2001/tag/2011/12/evolution/Registries.html [recorded in http://www.w3.org/2001/tag/2012/02/02-minutes#action01]
<trackbot> Created ACTION-667 - Check, when publishing and linking wraps, whether it's time to reinvest in http://www.w3.org/2001/tag/2011/12/evolution/Registries.html [on Noah Mendelsohn - due 2012-02-09].
<noah> ACTION-667 Due 2012-04-01
<trackbot> ACTION-667 Check, when publishing and linking wraps, whether it's time to reinvest in http://www.w3.org/2001/tag/2011/12/evolution/Registries.html due date now 2012-04-01
<noah> ACTION-595?
<trackbot> ACTION-595 -- Larry Masinter to draft a report on Mime and the Web -- due 2012-01-24 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/595
noah: Your thoughts, Larry?
Larry: mime provides a way of giving a persistent name for a language, which is different from a persistent name for a specification
... it's another issue where I think I understand it, but I don't know what we as a group want to say about it
... I get the impression that other TAG members aren't really into this
... that it's on the periphery
timbl: I don't think it's on the periphery, I think that mime is an important keystone
... other issues such as Javascript modules, and RDFa and HTML, are more topical
<Zakim> noah, you wanted to talk about worries about versioning
<Larry> should we wait until we're done with registries to tackle mime?
noah: I'm interested, but I'm worried that it leads into territory that the TAG has shown itself to not be good at closing on
... I thought the specific thing about the registration of media types where the specifications evolve
... lots of people think they know how it works, but we all think about it differently
... it's not lack of interest, just a sense that I'm not confident that we could get there
<Larry> maybe 'versioning' has become a bogeyman for the TAG? aversion to versioning
noah: if we could do something in 2-4 pages that was a how-to, then that would be great work
... but I think we'd thrash on it
jar: I'd like the more scientific approach rather than the prescriptive approach
... I think someone should write it down
noah: that's why I thought a wiki would be a good space to start
... but we didn't get to an agreement on how to do that
... the default is to close it and walk away
Larry: one thing to do, once we've done registries, we could take another task on the mime registry in particular
... it's the media type registry that's of interest here
<Larry> well, and the charset registry
<Larry> and the 'willful violation' issues around registries whose values are ignored for some purposes
noah: minute that we might want to revisit the issues of mime and mime registration after the registries work is done
Larry: yes, add media type and charset registries to the reminder
<Larry> there's http://www.w3.org/2001/tag/2011/12/evolution/MIME.html
noah: why don't you take responsibility to make sure that's included in scope when we look at registries
Larry: there is a document on MIME in particular (linked here)
... which points to the registry document, and raises specific issues around what MIME tries to name, and polyglot/multi-views etc
... some of the issues that would have to be done in a full exposition of MIME and the web
... so this document is a draft of what such a report might look like
noah: I thought the point was to not do the MIME-specific stuff, except in so far as it informs the general story
<noah> ACTION-595?
<trackbot> ACTION-595 -- Larry Masinter to draft a report on Mime and the Web -- due 2012-01-24 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/595
noah: Right, and we're proposing to put that off until at least after the general registries work is done, right?
Larry: I've added a link in that action item to the report
... so the action item is done
noah: usually a draft means that we're going to carry it forward, but I think you're saying put that draft down for now
Larry: yes, until someone else wants to do the work on turning that into a TAG document
noah: is there anyone else who wants to pick this up?
<Larry> and perhaps this is to just serialize this work so that the tag can take up things one at a time
Larry: basically there's a big raft of work and we need to focus on one thing at a time
... taking it up after doing registries is fine
noah: how should we track this?
<noah> close ACTION-595
<trackbot> ACTION-595 Draft a report on Mime and the Web closed
JeniT: suggest long-term action
<noah> ACTION: Noah to eventually, probably after registries work is done, ask whether TAG wants to reinvest in MIME/Web Due 2013-01-31 [recorded in http://www.w3.org/2001/tag/2012/02/02-minutes#action02]
<trackbot> Created ACTION-668 - Eventually, probably after registries work is done, ask whether TAG wants to reinvest in MIME/Web Due 2013-01-31 [on Noah Mendelsohn - due 2012-02-09].
<Larry> might put http://www.w3.org/2001/tag/2011/12/evolution/MIME.html in -668 as a starting point
<noah> ACTION-668 Due 2013-01-31
<trackbot> ACTION-668 Eventually, probably after registries work is done, ask whether TAG wants to reinvest in MIME/Web Due 2013-01-31 due date now 2013-01-31
<noah> ACTION-636?
<trackbot> ACTION-636 -- Larry Masinter to update product page for Mime and the Web -- due 2012-01-17 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/636
noah: purpose of project pages is to set down goals/dates/success criteria on big work
... we have draft pages on mime work
... my impression is that it doesn't tell the story of the first round of work, about the IETF draft
... I think it would be good to include that, and talk about what future work we might do
... such as registries
<scribe> ScribeNick: jar
<Larry> the product page depends on what others are willing to do, which i didn't have a good view
Larry: the product page is a commitment, so since there is no commitment - well, let's turn it into a wrapup page.
noah: Briefly tell the story of the successful work that happened
... Say we considered another round, and this is just on hold
<Larry> action-636
<Larry> action-636?
<trackbot> ACTION-636 -- Larry Masinter to update product page for Mime and the Web -- due 2012-01-17 -- OPEN
<trackbot> http://www.w3.org/2001/tag/group/track/actions/636
<noah> ACTION-636 Due 2012-02-03
<trackbot> ACTION-636 Update product page for Mime and the Web due date now 2012-02-03
noah: thank you
<Larry> ok
<Larry> action-350?
<trackbot> ACTION-350 -- Larry Masinter to revise http://lists.w3.org/Archives/Public/www-tag/2009Oct/0075.html based on feedback on www-tag and the feedback from TAG f2f 2009-12-09 discussion -- due 2011-11-29 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/350
<noah> ACTION-350?
<trackbot> ACTION-350 -- Larry Masinter to revise http://lists.w3.org/Archives/Public/www-tag/2009Oct/0075.html based on feedback on www-tag and the feedback from TAG f2f 2009-12-09 discussion -- due 2011-11-29 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/350
<noah> "Best practice for referring to specifications which may update" http://lists.w3.org/Archives/Public/www-tag/2009Oct/0075.html
noah: Henry had written ...
... Larry took action 350, where he came up with a different approach
Larry: I can't say it's an alternate proposal - I just didn't like where it was going
<Larry> i didn't really think i had something 'better' except a direction
<noah> I put Larry's proposed test in an e-mail: http://lists.w3.org/Archives/Public/www-tag/2012Jan/0043.html
noah: some people liked Larry's writeup more than others. left off with discussion of QA group's rec
Larry: action started in 2009. we talked, investigated. I suggested closing.
<Larry> http://www.w3.org/TR/qaframe-spec/#ref-define-practice
<noah> Thanks Larry
ht: I need to review in more detail
... what I wrote was what I understood to be practice that several WGs have been following
... so those wgs were not following the QA advice -or maybe I did a poor job of reconstructing their practices?
... I can't contribute to a decision since I haven't formed an opinion
Larry: ht to take an action?
ht: I'd only get to it this summer, but will do it if you like. Else you can close it.
noah: Preferences? Worth waiting that long?
ht: sure, but understand that there's no rush, sodue maybe this summer.
<noah> ACTION: Henry to review http://www.w3.org/TR/qaframe-spec/#ref-define-practice and see whether TAG needs to do more on references to evolving specs Due: 2012-08-01 [recorded in http://www.w3.org/2001/tag/2012/02/02-minutes#action03]
<trackbot> Created ACTION-669 - Review http://www.w3.org/TR/qaframe-spec/#ref-define-practice and see whether TAG needs to do more on references to evolving specs Due: 2012-08-01 [on Henry Thompson - due 2012-02-09].
<noah> close ACTION-530
<trackbot> ACTION-530 Draft slides for IETF meeting, with help from Larry Due 2011-02-22 closed
<noah> ACTION-644?
<trackbot> ACTION-644 -- Larry Masinter to draft proposed alternative text to e-mail announcing end of "product" work on HTML 5 last call ( https://lists.w3.org/Archives/Member/tag/2011Dec/0041.html ) Due 2012-01-10 -- due 2012-01-10 -- PENDINGREVIEW
<trackbot> http://www.w3.org/2001/tag/group/track/actions/644
noah: I wrote draft email announcement, LM revised it
... I'm fine with LM's
http://www.w3.org/2001/tag/products/html5review.html ?
<noah> Draft page from Larry: http://www.w3.org/2001/tag/products/html5review
<noah> http://www.w3.org/2001/tag/products/html5review-2011-12-21.html
<Larry> you already did that, Noah
<noah> close ACTION-644
<trackbot> ACTION-644 Draft proposed alternative text to e-mail announcing end of "product" work on HTML 5 last call ( https://lists.w3.org/Archives/Member/tag/2011Dec/0041.html ) Due 2012-01-10 closed
noah: all done here.
<Larry> this was done a long time ago, we just overlooked closing the action
<noah> Please look at: http://www.w3.org/2001/tag/products/index-2012-01-12.html
noah: No longer a priority product, OK?
<Larry> we'll do that after TAG reviews the final product page
Larry: Let me finish product page and get review before dropping priority?
RESOLUTION: The TAG is closing its work on MIME/Web, noting successful completion of 1st round of work. Product priority list to be updated after closing product page approved.
<noah> ACTION: Noah to update product priority list to mark MIMEWeb completed after final product page available Due 2012-03-01 [recorded in http://www.w3.org/2001/tag/2012/02/02-minutes#action04]
<trackbot> Created ACTION-670 - Update product priority list to mark MIMEWeb completed after final product page available Due 2012-03-01 [on Noah Mendelsohn - due 2012-02-09].
<noah> http://www.w3.org/2001/tag/group/track/actions/pendingreview?sort=owner
<Yves> http://lists.w3.org/Archives/Public/www-tag/2012Feb/0021.html
<darobin> http://www.w3.org/2001/tag/products/apiminimization-2012-02-02.html
ht: Maybe discuss e-mail from robin http://lists.w3.org/Archives/Public/www-tag/2012Feb/0021.html
<noah> Product page draft is at: http://www.w3.org/2001/tag/products/apiminimization-2012-02-02.html
<Zakim> noah, you wanted to make a chair's comment on scope
Robin: Broadened scope [see cover email linked from minutes]
<darobin> just a little bit of background on fingerprinting if it helps: http://panopticlick.eff.org/
Robin: ... that would make the finding more coherent.
noah: I'd prefer to get this little piece out sooner, then broader thing as followon
Robin: I understand, but I''m fairly confident that (a) can draft before next f2f, (b) problem with only minimization is that it is not timely - not relevant to APIs currently be designed - fingerprinting is a priority
... designs of APIs have changed since we started looking at this
... not a huge difference in workload to tackle entire thing instead of just part of it
noah: There's a bad history behind this general approach.
<Zakim> jar, you wanted to push back a tiny bit
<Larry> there's more to API minimization than what we're talking about
jar: I don't disagree with any of that, but... it seems to me fingerprinting is different from minimization. Not convinced it has to do with privacy. Not sure others agree. I tend to feel things that are mainly technical in practice feel more like security than privacy.
Robin: I somewhat agree. minimization is on the fence between security and privacy
<Larry> for example, geopriv in IETF took a policy approach that all locale information should be accompanied by an minimal expiration policy and a distribution policy . API minimization is only one part
<Larry> http://tools.ietf.org/html/draft-ietf-geopriv-policy-25
noah: Continue or take it to e-mail...this was an unplanned discussion
jar: We have 15 mins
Larry: I've been worried about the TAG pulling on a thread in this area... there's a huge body of work on this
... don't know how to part without getting sucked into something much bigger
<Larry> http://www.w3.org/2009/policy-ws/papers/Tschofenig.pdf
Robin: Agree. can't extract API minimisation from privacy context
<Larry> http://www.ietf.org/rfc/rfc3693.txt
noah: What I assumed scope was narrow, about granularity of requests -> might help privacy goals. People disagreed
Robin: How to present to users choices (in UI) about what info to give
... Very similar to fingerprinting
... user gets to pick which camera to expose
<Larry> look at RFC 3693 section 1 for a broader set of requirements
noah: Min. was cast as a JS API thing, fingerprinting was about what was on the wire?
Robin: You can do some MITM fingerprinting, but that doesn't go as far as JS-based fingerprinting
... Don't want to discuss every possible kind of fingerprinting, only scope is JS APIs
... I forgot there were other kinds of APIs
noah: I'm OK with doing something broader, but keep your eyes open, don't thrash
<Larry> I think this touches on too much and that a TAG document in this area without addressing the broad scope is really unlikely to be constructive to the privacy community
<Larry> -1
<darobin> +1 :)
Larry: Field is large, complex , many people working on it, TAG contribution unlikely to be helpful to people working on the area
Robin: To be clear, the primary customer is not the privacy community, it's the API design community
Larry: Motivation for talking about APIs - there are lots of guidelines for API design, we don't want to get into that business either - too big similarly
... another community to integrate with
<Larry> API design guidelines for web APIs?
<noah> I agree with Larry in the following sense: if we're going to say that minimization is related to fingerprinting...well...anything we do with API design is going to drag us into the many broader issues of good API design.
Robin: There are lots of guidelines, but very few for Web APIs, they are sorely missing, this is a real need
<timbl> I think that Dan A may say that it would be worth just having the API minimization done say just that before doing something larger
<Larry> so the client side storage finding might recommend API design?
Robin: very useful, a community we need to engage with. This is a good start, test the waters
... I propose, let me start on broader scope, then retreat if necessary
<Larry> i don't think it's foolish, it's more a matter of setting context
noah: I hear some concern APIs get us in hot water
<Larry> I think we need a framework for understanding how the work fits into webarch and other things
timbl: If Dan were on the call, he'd advocate for API minimization *because* it's limited... that was the approach he favored for TAG publications, small
... later aggregation works
<noah> Would it make sense to do a product page which is, for a few weeks, not settled on scope, and sets April as a decision time on scope?
Robin: My problem - I don't think this is such a big broadening of scope. Let me clarify in followup. Dan & I have talked to other groups, and they don't really care, too fuzzy, whereas fingerprinting comes up all the time
<noah> I'm personally not that enamored of minimization, but the question is only partly whether other people like it: the question is whether there's a sound architectural principle they should learn
<Larry> i'm still not clear about whether this is 'good practices for API design' in general, or 'good practices for Privacy' or something else.
<darobin> Larry, the former
noah: (listing options)
<Larry> design principles and recommendations need to be in the context of a problem they're solving
adjourned.