Technical Plenary Day 2013

13 November 2013

These minutes are public


Jeff Jaffe greets audience
  1. Welcome
  2. Prof. Jinpeng Huai
  3. Dowson Tong
  4. Normative References
  5. The Document License Experiment in the HTML Working Group
  6. AB's W3C Process document review
  7. Discussion with the TAG
  8. Open Mic session
  9. Setting up breakout sessions

Scribes: Phil Archer, Michael Cooper, Ivan Herman, Philippe Le Hegaret, Amy van der Hiel

See also: IRC log | Member-only minutes of Advisory Committee Meeting (Day 1, Day 2)


Tim Berners-Lee speaking

<PhilA> Wu Yi Huan (Vice-Mayor of Shenzhen) arrives, introduced to TimBL

<PhilA> scribe: PhilA

<scribe> scribeNick: PhilA

jeff: Welcome everyone

[Jeff thanks Tencent for their sponsorship, coordination and help in bringing TPAC to China]

<paulc> is Paul Cotton, HTML WG co-chair

Jeff: Also thanks to the city of Shenzhen and our Beihang Host in Beijing
... we'll have a brief welcome from each of those organisations

WuYiHuan: On behalf of the mayor and the city of Shenzhen, welcome
... Shenzhen is on the south coast of China - a young city in a pleasant environment. We have won many awards, incl. city of design. Known for piano, golf and libraries
... As part of opening up, welcomes competition and more
... Over 3 decades, many start ups have begun and gone global, incl. Tencent
... We are striving to achieve sustainable socio-economic growth. Shenzhen broke more than 200 bn USD in business
... This is the first TPAC outside the US/EU - Shenzhen was a good choice
... This meeting will, I believe, be a boost to the ICT industries in the Pearl River Delta area. Will advance Shenzhen's work in standardisation globally
... Finally, I'd like to wish the meeting a great success


jeff: Prof. Jinpeng Huai is the person I've worked with closely over the years on opening the Beihang Host

Prof. Jinpeng Huai

JinpengHuai: On behalf of Beihang host, welcome to TPAC. First time this event has been held in China
... Shenzhen is a showpiece of China's opening up. Now among the most far developing cities in China. Home of China's most successful hi-tech companies
... On its way to becoming the biggest online market
... This week, a 24 hour online shopping carnival - many billions of US dollars spent
... The Web platform includes increasing numbers of Chinese-origin ideas
... Open Data opens many possibilities
... Beihang became first host of W3C in China - helping industry in Asia Pacific build the Web
... We really appreciate the generosity of Tencent - I wish you a successful TPAC here in Shenzhen

Jeff: Prof Huai and I have both praised Tencent... so welcome Dowson Tong

Dowson Tong

DowsonTong: As exclusive host for W3C TPAC2013, honoured to welcome you to Shenzhen
... W3C continues to drive the development of the Web. Tencent joined 2 years ago to make our contribution to this
... 15 years ago, Tencent was founded here. Since then we have grown a great deal. QQ (browser/messenger) is the most used in China
... Our services that Chinese companies use depend on the open standards of the Web
... that's why we're here
... finally - wish you success for TPAC and enjoy your stay in Shenzhen

Jeff: The Tech Plenary focuses on technical developments. We have a morning full of presentations on the tech developments in W3C.

Normative References

Jeff: It's a question that comes up - how and when can W3C specs refer to external docs normatively

Plh: Document available at

<dom> Slides on Normative References

Ralph: We wanted to talk about changes we've made to make transition easier

Slide 1

<dom> Criteria used by the Director to evaluate normative references

Ralph: The existing policy has worked well but it's become a problem in recent years. If we needed to refer to something normatively, required whole external doc to be of similar status, including non-W3C standards
... The director, often delegated to Philippe and me, looks at the normative references (slide 2)

Slide 3

<dsinger> the document doesn't seem to address the stability of the 'pointer' itself; i.e. given the reference text in the W3C document, can the intended referenced text actually be found over time? ideally, the document has a stable URL

Ralph: Grouped what we think of as important into 3 groups

Slide 4

Ralph: Ian Jacobs has written about this in relation how others should refer to our documents. That's out of scope for our work - we're just talking about references made within our own docs

plh: Slide 5
... Third bullet is more fine grained. How stable is the referred to element within a doc? It might be stable although others around it might not be.
... Are you referring to a feature or just a name? Often these days a spec references a definition so it doesn't need to rely on the details
... slide 6

<dsinger> nor does the text establish that the document can in fact be acquired 'reasonably' (e.g. you don't have to sign a license just to get the document, or join another body, or the like).

plh: Market may have a view on stability and need for progress
... slide 7
... slide 8 which refers to

Ralph_: We have a moment for questions

David Singer: Can you find the thing referred to? It may move (not everyone keeps URLs persistently).

Ralph_: We've not addressed that, no, but we have addressed the other point you raised about whether the doc is available generally (not behind a paywall/membership)

Jeff: Thanks the Shenzhen officials for coming

The Document License Experiment in the HTML Working Group

<MichaelC> scribe: MichaelC

Timbl: these are exciting bits of technology this morning aren't they? :)
... but they're important topics
... if you get document license wrong, other things go wrong
... regarding question of reference going 404, could request permission that we host a copy
... in a report recently 45% of US supreme court decision references are 404

Timbl: on to licensing
... there has been a lot of discussion about what license we should have on our spec
... not resolved
.. some don't want specs to fork because it impairs implementation
... others want to be able to use anything anywhere
... for HTML we have gone ahead with an alternate license
... cc-by Creative Commons With Attribution
... anybody can republish but must acknowledge source
... since the time we made the decision to adopt that license for that spec


<PhilA> I see your 2.0 amy and raise you to 3.0 :-)

Timbl: desire to be able to release under GPL has been expressed
... viral license, anybody can republish indefinitely
... under cc-by someone can republish but they must credit
... it's an open question of whether someone republishing in accordance with cc-by has right to put their copy under GPL
... (legal interpretation question)
... the question of whether Consortium should go this direction is open

Timbl: my personal view is copyright is not the tool to use to avoid forking
... even though in fact we want to avoid forks
... the right to fork keeps us honest
... we have to act in such a manner that people don't want to

Timbl: my view on the legalities is that if GPL can't be used on a cc-by document, then there is a bug in the GPL license
... there should be clear lines between the licenses
... I would have thought that GPL and cc-by would be in sub relationship to each other

Timbl: Yesterday's AC meeting was a cozy space, now we're in a big room
... please pretend it's a cozy space for the upcoming discussion

<invites discussion>

<ArtB> Is there something more to this discussion than Taste Great vs. Less Filling?

Manu Sporny: you must have considered public domain license at some point
... there seems to be personal desire for public domain
... why is it not considered more strongly?

<ArtB> It seems like the minute W3C put HTML5 on github, this whole discussion about copyright became moot

<ArtB> What am I missing?

Timbl: I wasn't aware of a movement towards public domain desire
... perhaps have not wanted to support ability to remove attribute and provenance
... it's a question of winning over hearts and minds
... would like information about valid use cases in which preserving attribution is an impediment

<dsinger> cc-by is 'close to public domain'. wanting attribution doesn't seem unreasonable

<ArtB> copyright isn't going to stop spec bifurcation

<ArtB> it's here, we have to deal with it

David Baron: cc-by isn't just cc-0 plus attribution
... there are additional requirements
... think the incompatbility with GPL comes from those additional requirements

<dsinger> also, cc-by satisfies the use-cases and requirements we collected. we can't move the goalposts

<ArtB> I don't see how copyright is going to help

<ArtB> (IANAL ;-))

Timbl: I imagine so

Larry Masinter: though the concern came from organizations that didn't want IP released via an open license in W3C spec
... so the license binds participants together

<ArtB> putting something on github says "please feel free to fork me"

Doug Schepers: I share the personal preference for cc-by
... I agree with attribution
... also think doc should be as available as possible
... I work with on documentation
... have trouble pulling in images
... with cc-by could pull spec content into documentation more freely

<chaals1> [ArtB, the one place copyright helps is when some large industrial consortium takes the spec and produces a b0rken fork that they inflict on us. Since they think in terms of copyright and lawyers, they are in fact vulnerable to being told off through a notice regarding copyright. Against which, it can be annoying in other places...]

Timbl: so if there is example text under W3C license, can't be pulled in?

Doug Schepers: because web platform docs is cc-by, can't pull in the content under the more restrictive license

Ian Jacobs: there are a lot of people here, maybe we could poll the room

Ian Jacobs at the microphone

Timbl: hmm, we didn't set up for a valid poll, but could try humming
... not binding, it's a bit random
... let's consider four options: W3C license at is now, cc-by, cc-0, don't know/ don't care

[W3C license - almost indistinguishable support]

[cc-by - modest support]

[cc-0 - roughly the same, maybe slightly more]

<scribe interpretation of hum poll>

[don't know/ don't care - also modest]

<chaals> [scribe seemed to hear something different to what was over here... conclusion, we need to learn to use the hum better, or should stick to the clap which is more usual here]

Glen Adams: why can't content be referenced via link, why does it have to be copied in?

David Singer: you could ask W3C for permission to repurpose, answer probably would be yes

Doug Schepers: there is just enough confusion here that mistakes could be made

David Singer: could use reasonable portions of spec, seems clear

Doug Schepers: communicating all that difficult
... anyway, want to solve the general use case, not just mine
... if we want specs to be used as widely and interoperably as possible, let's open up the license

Manu Sporny: there are instances where attribution requirements don't make sense
... e.g., primer documents that incorporate content
... can we give WGs ability to decide which license they want to use?
... know some groups would have rather published as public domain
... fear of forking, using spec to competitive advantage doesn't seem like strong argument
... people know W3C is authoritative source
... seems the fears are driven by legal departments
... so could allow groups to choose between cc-by and cc-0

Timbl good point, though if groups have that option, they'll spend time debating license rather than developing tech
... we have granted license for republishing before but didn't allow changes, we retain change control
... otherwise you have a destructive fork
... two versions of web

<masinter> who's interested in a forking policy?

Mike Champion: fundamental value of W3C is consensus and patent commitment
... we resist change because concern that forks without attribution could result in specs that appear to be e.g., HTML but lack the patent commitments
... yes lawyers are speaking to that but don't think it's just a lawyer POV
... let's continue experiment with cc-by and come back in a year or two to revisit

<dsinger> by the way, it's clear that the RF grants from members extend to the W3C spec., not to possible uses of the same text/ideas elsewhere. "You may not have a license" is an appropriate warning...

<cwilso> ...and about 95 other people. :)

AB's W3C Process document review

Steve Zilles presenting

<koalie> scribe: Ivan

<koalie> scribenick: ivan

<dom> Steve's slides

Steve Zilles: I am currently chair of the process document revision group, Charles is the editor of that new document

<ArtB> Is this the doc ?

SteveZ: what I want to do is to give a quick overview on why we are here, what we are proposing to change
... it is currently in last call review
... What we started up with a process 'waterfall development'

SteveZ: starting from one step, refined, then I have functional completeness, then sent to last call, then to implementation experience, finally a rec
... what we discovered is that this is not what is happening
... in software development, or standards' space either

SteveZ: more accurately what is happening is that the fpwd may not have all the pieces,
... some of the pieces may be relatively developed (eg, coming from a CG)
... I have a series of drafts
... some of the pieces may be implemented relatively early
... ie, getting to CR i may have a number of implementations and tests
... instead of having a sequence of steps I may have parallel pieces
... some of them may be stable enough for reference
... what we try to work for is doing the specs in a more agile way
... doing them in smaller chunks (modules, fragments)
... with implementations and (experimental) developments along
... finally developing tests in conjunction with the spec development
... this is already being done without a process change
... We began to look at this issue in '11, asking people about their problems

SteveZ: the AB with the people from those sessions, sent a compiled list to the chairs
... 'which of these are important to change'?
... Of the 12 priority items, 6 are apply to chapter 7 draft we are talking about here

1. integrating implementation in the process

2. process does not match modern developments

3. desire for modern references

4. last call may not be as useful as intended

5. interaction between lc and cr is confusing for outsider

6. lack of test case and interop development

SteveZ: Last year at TPAC we reported, and then we were looking at these issues
... we realized that we can make a number of process change
... eg, we could get rid of activities
... eg, we could break the document
... but applying our own rules on agile dev., and focus on one piece
... most of the things are concerned with the interaction of lc, cr, and pr
... we want to get rid of unnecessary steps
... that led us cutting down our work
... in future we will make the other changes, later

SteveZ: What are the changes?
... No change of the patent policy
... we changed the role of last call, separating the pr exclusion issue, and is now attached to cr
... the other was to get the responsibility of the wg to get large review
... we simplified the diagrams to fewer maturity levels: wd-s, cr, and rec
... doing parallelization we overlap the ac review with the cr

SteveZ: Another piece was to clarify and simplify the document
... removing the non-normative advice, and move that to a separate document maintained by the team
... we made it clear what the normative pieces are

(Steve shows extracts of the document)

SteveZ: we try to list what _must_ happen
... in the same process we identify _may_ and _should_ requirements
... We also try to identify what the director considers adequate review and implementation experience
... guidelines and not rules

SteveZ: Finally, we require public statemetns on the director's decision
... We already identified issues
... I refer to last call candidate rec for ipr text issues
... so we may call that cr
... and add an explanatory text for ipr

SteveZ: Early in the process things can get fully defined before the doc as a whole
... we encourage early wide review
... documents identify sections in the document that are available for review early in the process
... issue is to signal that the document as a whole is complete
... There is also an issue what should happen in the status section; today it is more boilerplate and that could be removed
... The wide review piece of lc does not require director's call
... It was suggested to rename 'rec' call 'standards'

SteveZ: The LC [of proposed revised Chapter 7] is open until nov 27, comments to
... issues are public
... irc is public #w3process
... minutes also public

SteveZ: Target to have a final version for AC review in January

SteveZ: Questions?


<dsinger> is there a reason this isn't on My Questionnaires?

<dom> dsinger, I think this is a call for comments, not the formal review yet (?)

<jeff> [Fantasai: Applause! Awesome!]

Judy Brewer: I think there are good changes; one particular issue
... there are a few groups that have a horizontal functions, like WAI PF, they have to schedule very carefully to get the expertise available
... I noticed to have a heads-up for next steps

<dino> Feedback to organisers -- I'm not sure this presentation is the best use of our collective time. I'm pretty sure most people are not paying attention, and it could have been discussed on email. (Also, even the presentation makes it clear this is for AC review). I'm not sure why it wasn't a BOF.

Judy: but cross review groups would need a heads-up for the heads-up
... is there a possibility to provide more assurance that there would be a heads-up (not just recommended)
... and whether there should be a specific practice for all the groups to use

<jeff> Dino: Thanks for the feedback. The theory (maybe wrong) is that developers are impacted by the W3C Process and we want to make sure they have awareness and we get their feedback.

<ArtB> Good comment Judy!

<ArtB> I've been wondering the same thing

<ArtB> How many horizontal groups do we have now?

<ArtB> Security, Privacy, A11Y, TAG, I18N, Mobile, ...

SteveZ: the category of 'other things' includes a provision to change how things are chartered, also with the provision to the horizontal groups, opening a liaison experience to have an identified person for a continuous information flow

<dino> jeff, Oh, I agree that it is important and that WG members should give feedback. I just feel the collective time of 300+ people is extremely valuable. I might be in the minority.

SteveZ: the second one is: in some cases a major arch change late in the process is not the good place to do it, so each release should identify what needs review

charles: one of the goals was to get out of the way of people doing the work
... you cannot get into cr unless you got the right review
... but we do not tell you how to do that

<jeff> Dino: Fair enough. In fact that's why most of the "plenary" is breakouts. For a variety of reasons, we needed more "plenary" this year.

charles: having a process setting rules about scheduling did not seem to be helpful
... you got to work with the dependencies, other groups, etc.

<ArtB> So, are we now going to need a process to manage the [review] process?

charles: if you get it right, and normal practice to coordinate with other groups is go talk to them, no nasty surprises, then it works

<dino> jeff: the breakouts start at 2pm right? That's not "most". Anyway, I don't want to complain. I understand there are important topics that need to be broadcast to a collective audience.

judy: i would encourage regular mechanism

<ArtB> +1 to Judy's proposal

Paul Cotton: the normal wg transition to the old lc the group did of its own, and cr was with the dirctor, when formal objections are dealt with

charles: any time you go to the director you get the formal objections are handled, currently lccr

<ArtB> Does the AB have some data that shows that if a group had followed this proposal it would have made a substantive difference?

SteveZ: that is when this _have_ to take place, although you can discuss with the director earlier

Paul Cotton: postponing formal objection is actually a good things, people can calm down...

Phil Archer: you mentioned the 'going back', where do you go back to

<fantasai> ArtB, csswg would not delay publishing updates to CRs in order to batch them if it were possible to edit a CR without changing its status to WD

charles: the normal situation you go from CR to a new CR

<ArtB> but if "editing a CR" mean adding any new features or norm refs, then a Patent exclusion must be done

charles: today the situation is to go back to a new lc, pattern exclusion, 3 weeks, etc, before they are allowed back to CR
... the public sees 'last call' and another 'last call', etc, and that is confusing
... we tried to eliminate that, pushing wg-s to get their documents reviewed right, and minimize the confusing symbols even when going back

SteveZ: thank you all

<mchampion> Speaking as an AB member who has been in these discussions: I'd just reiterate that the spirit of these changes is to tell WGs WHAT they need to do before asking to go to Recommendation, not so much HOW they need to do it. Maybe some "how" things need to be more explicit, but there will be a tendency to say "let's have WGs and WAI/I18N/etc figure out how to optimize their own interactions, not suggest that the process doc cover all contingencies.

Discussion with the TAG

TAG Members on stage: Peter Linss, Alex Russell, Tim Berners-Lee, Anne van Kesteren

<plh> scribe: plh

<koalie> scribe: Philippe

<koalie> scribenick: plh

[Dan is trying to figure out the architecture of a VGA plug]

<Ralph> [to Dan's defense, this VGA connector is not quite standards-compliant]

Dan Appelquist: the idea is to have a Q&A
... so what is the TAG?
... it's a special WG chartered by the W3C Process
... to document and build consensus around the principles of Web Architecture
... resolve arch issues
... we have Anne van Kesteren, TimBL, Alex Rusel, Peter Linss and Dan Appelquist on stage
... (missing are Jeni Tennison, Yves Lafon, Sergey Konstantinov, Yehuda Katz, Henry Thompson)

<Ralph> Technical Architecture Group home

Dan: we're focusing on web of applications
... we give reviews and feedback
... for example the web audio, webrtc, web components, web animations, web crypto
... helping with liaisons, like IETF and TC39
... JS, JSON
... best practices guidelines
... EME

Dan: security of the Web
... looking into web dev outreach
... mostly oriented around Q&A and feedback
... next f2f is in London in January and in SF in early April
... looking at doing outreach there
... lastly, we have 2 seats up for election
... representatives on the TAG are elected

Dan: if you have questions about the election, come to the mic or us
... we'd like your feedback or/and questions
... we'd like to hear from all, especially WG participants

<ArtB> [we want feedback from _real_ people ;-)]

Dom: the work you guys started is really good
... re API design, that's a constant question
... Robin started an API cook book 2 years ago
... would be good to take it on and continue it

Alex: I support that for a long term goal
... but it seems more productive to work hand-in-hand with you on specific issues
... how it interoperates with the rest of the platform...
... trying to make your document better
... we'll continue the reviews
... and extract commonality afterwards
... so we support the goal of guidelines
... but trying to gather experience

Dom: agree with the approach
... but there are small things like how to write variables, etc.
... taking time from WGs

Alex: this tends to uncover large issues however

Dom: How do we deal with permissions and managing access to APIs?

<chaals> [+1 to Dom. Getting this stuff written down should be a key goal]

Alex: don't expect magic from us, but we're happy to help
... ping us and we'll schedule time
... we're a consulting group at this point

Dom: I'll bring this feedback back
... increasing the visibility of the work would be nice

Alex: we have limited bandwidth
... it does require study for us to come in

<virginie> +1 to dom request

Alex: would be great and as we continue to show success of improving specs, that profile will be raised through success

Dom: before the groups can invite you, they need to know that they have this ability

Manu: when I started to work at W3C, the standard was to avoid the TAG
... was meant that the TAG wasn't affecting the work
... I understand that the TAG wants the architecture to work, and I understand the recent changes in the TAG
... but when I hear things like limited bandwidth, that puts me back into the previous mindset
... if you have limited bandwidth, peharps there is an opportunity to create a spec reviewer task force
... or increasing the number of folks on the TAG
... the other part is that not many people know that you exist
... so maybe scheduling time whenever you see a new draft, trying to inject yourselves could help

Alex: we can't review all drafts in the world, but we'd like to engage with groups as early as possible
... to avoid pain as much as possible
... but I understand the critique
... we'll continue to build a library of successful reviews
... to help the visibility of the TAG
... we'd like to collaborate with you early
... and don't think that we can't help

Dan: +1. the TAG was helpful previously but I'd like to see that be the norm
... that working with the TAG is a positive experience

Junglee: re collaboration between w3c and whatwg
... I work on XHR
... recently Anne was defining of fetching resources in the whatwg
... how can I handle references to that particular spec in the next version of w3c xhr?
... same problem in the html wg
... can the TAG address this issue?

Anne: we didn't discuss this in the context of the TAG. it's an issue related to licensing and process
... the best people to ask is the AB or the AC

Jungkee: how about referecing the fetch spec?

Anne: you should just reference imho. no big deal to me.

Natasha Rooney at the microphone

Natasha: the TAG meetup event in london was awesome
... please do more of it
... maybe incorporate the TAG into chairs training as well?
... maybe a wiki page
... so the chairs can be the voice of the TAG and have the discussion when there is a need to

Alex: seems good

Dan: so it's all clear?

Timbl: yes, we could have done the chairs breakfast

<ArtB> what?

Timbl: the chairs training is an important idea
... the very first chairs meeting was fantastic
... difficult to get everybody together

Charles: the TAG as a technical group is elected
... that seems weird
... how does it really work, ie limited in size and by election?
... do we have the right people?
... does the limitation help?

Timbl: ask the judges that
... to the extent that folks are elected and could be rather random, I get along with everybody
... I forward to see who is going to be
... but at the moment, they seem reasonnable people
... it's good to have continuity
... as welll as changes
... variety
... and overlap

Dan: the last tag election when a number of new members came in
... and a lot of discussion in the AC
... and lot of effort
... to promote the idea a change of expertise
... that worked, even if it was political
... to me, that's democracy in action and was quite effective

Alex: it strikes me that it's about priority setting
... we could spend time dealing with issues that don't impact people
... we'd like to hear your priorities for sure

Mike Champion: TAG history question
... constrasting TAG and the IAB
... was told that IAB has less judgmental process
... more careful in balancing point of view and expertise

<dom> [I think Michael is alluding to the IESG]

Mike Champion: did you think of using that kind of model for the election?

<ArtB> Wouldn't it be a whole lot simpler and more inclusive to simply morph the TAG into an Interest Group?

Tim: different structure
... IESG have the area directors
... IETF doesn't have domain leads because it's all volunteers

<ArtB> Especially since it now mostly a Review + Education and Outreach group?

Tim: [...] the election is good to bring people fresh
... this time we had people blogging on what the TAG should do

<dom> [there is also the Internet Architecture Board (IAB) ]

Tim: was pretty useful process and valuable one
... doesn't happen when you have a community process
... so don't think that the TAG corresponds to anything in the IETF

Ann Bassetti: one bit of history, the AB started the TAG
... the technical questions used to come to the AB which wasn't the appropriate venue
... thus we created the TAG
... we purposely made it partially elected and nominated
... to balance the technical expertise and so on

Charles: one kind of things is struggling with bandwidth
... in specific area

Dan: security expertise is lacking in the TAG

Timbl: it's good to have javascript expertise
... was unfair to talk about web arch without it being written down
... so the TAG did that
... while being reactive
... the doc went down deep to answer questions
... while being light on some aspects

Alex: we're light on media graphic expertise
... interaction between codecs and network layers
... SSL
... that's a new thing for the web platform
... we're leaning on the expertise of others

Dan: and we need people who can write specs :)

Jeff: thank you

Open Mic session

<amy> scribe: amy

<koalie> scribenick: amy

Jeff: we've been asked for open mic. this is an informal time for questions, web, W3C
... generally Ann Bassetti helps facilitate for those who don't have english as first language. you can set private channel w/ Ann, grab her or talk to her.
... for those who would like to raise issues but might need help, we nominate Ann.


Ann Bassetti: I'd suggest that anyone who does not know how to use IRC, or go to for basic instructions. much of what gets discussed here is visible there

Jeff: as people slowly come to mic. closest to me we have Ralph Swick, COO; Philippe Le H├ęgaret, Interaction; TimBL, director; Ian Jacobs, MarComm; Philip Hoschka, Interaction; Judy Brewer, WAI; Wendy Seltzer, legal; Doug Schepers, Developers Relations.

Cameron McCormack: for a while now the SVG and CSS Working Groups have wanted to publish docs w/ styling. when can we do this?

Ian Jacobs: please come to site redesign session today. we'll show you examples. 50 pixel style ideas. wireframes don't extend to specs
... I do have style drafts, if you're interested in TR.
... if you're interested in sponsoring please talk to me

Cameron: quickly, what's the time?

Ian: that depends on answer for sponsorship (joke). goal is 2014

Manu: we had discussions of internet of things. economy of startups, kickstarter
... a lot of these startups are using web as core part of business. they're kickstarting a process. but if we look at bringing them here, to help w/ standardization, it's a bit confusing, at least to me, what the selling point is
... how do we say pay 8K get something in a few years.
... how do we get them involved, membership early in process, and then anything else management has been thinking about

jeff: great question. several answers

Tim: I'd like to point out we're up here, you're sitting there. I'd like to see answers as well. it's an asymmetric thing

jeff: great question. i think you have to break it down by use cases. first, a lot of startups are doing stuff that's not ready for rec track. we created Community Groups to get the visibility in w3c w/out overhead
... participation is free. no barriers.
... another answer to 8K: we made another class called start-up. it's 2k.
... if you want to, if it's important to business, not just building product but setting standard.

Manu: is that every 2 years?

Jeff: 2k per year for 2 years

Manu: start ups say not going to be able to continue after 2 years, not see work through
... they see the revenue through, if they want to really start something at w3c
... build something in standardization, only halfway through process

Jeff: again, it's a use case issue. it's not usual that startup are ready to be members. if they think that year three or four they're ready then they join in year 3, year 5 if they're optimistic it's not unreasonable.
... startups aren't one size fits all. we have several options

Tim: do you think we should have modified version of startups?

Robin: i wanted to address Manu's issue, having worked in startup, cost isn't membership it's traveling for mtgs. one thing we can do is keep reducing overhead of working on standards
... we're thinking of how to keep pushing at process level and culturally. bubble down to chairs, remove overhead, push burden away from community
... process we have is there for a reason. the world has changed a lot. a lot use git hub, put stuff up, we need to be able to match that

jeff: I think we all agree

Charles: I nearly agree. there's a lot of stuff on git hub that's junk. they end up disappearing because they're not good. we can't lose sight of idea we need to do good things, not just things fast.
... but reducing overhead is a good idea

Yosuke Funahashi at the microphone

Yosuke Funahashi: co chair web and tv. i have a question re: testing. we had a four year plan
... we started this last year. plans to publish results?

<Ian> [There will be a testing presentation at the AC meeting tomorrow morning]

PLH: Tobie Langel will share results tomorrow at AC. I think there's no reason to make it hidden. I'm sure he'll finish his slides in time :)
... we've gotten good results on testing. James Graham doing beautiful work
... there's still plenty of work to be done. I've been pleased w/ the work tobie has done. really looking forward to continuing that in next year

jeff: shall I conclude everything else in consortium is perfect?
... for those who didn't hear, fantasai said things are going in right direction

Alex: several years ago i talked about individual membership. I mollified at the time by idea that individuals could participate in cg process
... w/ CG now a few years old, ask how it's perceived and bring back indiv. membership

Ian: as far as how it's going, we heard good reports from membership at AC. more than 3250 participating in CG, RF commitments in CG
... some push specs to rec track. interest in how to do that. CG visibility, ideas of how to join. systems improvements. based on what i'm hearing, there's support. we have more to go

<koalie> [ There are currently 3374 people participating in community and business groups. ]

Jeff: briefly, individual membership - what problem to solve? to have participation? if we think we need them to pay $? possibly not worth it

Tim: I've always been interested in individual membership. people mail me, saying "I want to join because i think they're a good thing"
... but others say "I'm involved in X thing". so I think maybe we've found the overhead over actually running this wouldn't be paid for by incoming $
... we've gotten better at taking credit cards at meetings. from the community people say i am a member even if they don't participate
... could also be read-only membership
... be a member to see what's going on, out of interest or support, separate level where you've got an account where you can be allowed to read-write and
... I'd be happy to revisit this

Judy: in terms in community groups it's been interesting to see, exploration of different topics, at w3c it's allowed us to get new things. one of the hallmarks of w3c work is review
... way tries to help groups w/ accessibility. PF groups look across groups to see how they're performing and we bring to those in field
... right now we try to provide to w3c as a whole we're not able to do for CG. they don't have opportunity to work w/ us. if you're in a CG and you have suggestions of how we could be better connected, let us know. talk to Janina or Michael C
... first question had been about entrepreneurial groups, we see groups, doesn't matter small or large, some groups are small and dedicate people and then step up as they can

Jeff: thanks. we're running out of time. Ask for Alex clarification

Alex: a few issues to be sorted out. one issue is the professional affiliation question. they'd like to be publicly affiliated w/ W3C
... question of utility and the responsibilities it implies. one issue is we don't have surfficient representation from UI

<Ian> [Alex makes good point about end user representation in groups]

Alex: could elect people, we don't have strong sense of who represents users or who represents vendors
... we want both perspectives but we wear too many hats. if you have to have too many things in head, it's difficult

Jeff: this is a larger question. I'd observe that individual membership is professional affiliation, many organizations, closest might be Internet society, they have to be thoughtful about what services they provide
... they have to think about how they support chapters. it would have to be separate than what we do now - what resources, what we offer.

Manu: a regulatory policy question. payments. one thing coming up over and over again. what's legal to do in what country. what policy makers do we need to have involved
... w3c started to work in mobile. but i don't' see legal and regulatory expertise. we need this
... people like world bank, we need to have a much tighter dialogue. does no good to have a spec that can't be implemented. is there any suggestion that w3c has for this for web payments or web identity. where should these people be placed?

Jeff: great question. the current group we have that most resembles that is Tracking Protection. We've reached out to regulators, and involved in dialogue but I don't think we've created a repeatable process. we'll need to think about that more
... this needs to be on agency, and I see Rigo has come to the mic.

Rigo: the experience from the past, personal application of German minister of interior, how to link passport to commerce. I think getting in a permanent basis would be difficult. but for workshops, we have pretty good channels into governments
... we can get them into workshops, but binding them to WG would be difficult.

Daniel Austin: ebay/PayPal.
... some expertise about regulatory for payments would be part of what we do. we have some expertise. we can bring more of this, we suggest get more payment groups to participate in w3c
... problems exist for all payment guys. we'd like our competitors to participate as well

Jeff: in spring we plan on having a web payments workshop, Paris March 24/25
... we want to use it to reach out before getting them to join. if we can also use to bring in legislators that would be great

Jeff: thanks to all for your questions and to w3m for addressing them
... the afternoon as in several years will be community breakout sessions. we have several groups in play.
... Ian will explain what will happen over the next few hours.

Setting up breakout sessions

Crowd setting up breakout session topics

<koalie> [setting up breakouts]

Ian: we're doing a raffle and giving away pairs of penguins. put your business card in the box
... or your name on a piece of paper. we're not using data for anything except wedding penguins

Ian: between now and lunch we want you to surge to front of room and fight for your position for a breakout, there are limited slots
... currently 25 proposed sessions

<dom> suggestions for TPAC break outs

Ian: if we run beyond thirty, we can have brawls to set space.
... we don't have post it notes, we have blank pieces and pens.
... you will write session name and your name.
... some like to get their idea in, others wait, and we set up a grid
... you'll seed the schedule and break out session

<cwilso> explicit request that someone echo the schedule planning here in IRC

<dom> cwilso, the schedule will be reflected in the wiki


<cwilso> then just a notification when the schedule times are set in the wiki

Ian: in printed program and in tpac wiki there's guidance for good practice for running breakout session
... e.g.:
explain why there. it lets people decide to stay or not. what you expect to accomplish. use the wiki to take notes
... e.g.:
i've added site redesign. make your own irc channel and let us know in the wiki.
... e.g.:
find a scribe, remember not everyone speaks the same language. notes helpful.

Ian: 15 mins between slots should give travel time.
... please try to start and end on time.
... we'll put up grid on line after it's stabilized.
... those will be public.

[scribing ends]

<cwilso> Public request that the Music/Web MIDI session make it early

<olivier> cwilso, the midi session is likely to be the first one after lunch FWIW

<cwilso> Yay!

<cwilso> eeeeeeexcellent.

<cwilso> that's in approximately 1.5 hours, correct?

People reporting on their breakout sessions

<olivier> aye

<cwilso> thanks!

<richt> If you're attending the 'Web of Things' breakout session please join the #webofthings channel

<cwilso> Web MIDI chat is on #webmidi

<ArtB> and Web Payments' channel is ...

<ArtB> #webpayments

<ArtB> who wooda' thunk


<koalie> [breakout reports in the main room]

<virginie> FYI : security breakout session minutes are available here :

<koalie> Some breakouts minutes already linked from the breakouts grid: