TPAC 2011:

Minutes of TPAC Plenary

Contents:

  1. Successes and Challenges
  2. Web and TV IG HTML 5 Proposals
  3. web content interoperability
  4. Agenda building
  5. Sharing and Advocacy Plenary
  6. Wrap up

See also: agenda · IRC log


Crowd Shot

Full room during plenary day

<matt> scribe: Matt

Jeff: Today will be breakout sessions mostly.

<FabGandon> Venue of the TPAC next year : http://www.ccc-lyon.com/home

Jeff: Two discussions today, one on Web and TV and another on Interoperability.


Successes and Challenges

Jeff Jaffe

Jeff Jaffe

Jeff: In thinking about what to present to this group, I spent a lot of time sharing a focus that we have within W3C on the business model of the W3C.
... I wanted to do something different, as many of you have heard about the business model for quite some time. Plus the TP audience is mostly the innovators, who are more interested in how we create standards at W3C.
... You all see pieces of that with your team contacts and WGs, but many of you haven't seen the broad process.
... I don't have all of the answers here. I have some answers, some open questions, but we want your feedback.

<inserted> [ Jeff's Slides -- http://www.w3.org/2011/Talks/jj-tpac-20111102.pdf ]

Jeff: W3C Organization Focus, just two slides on this.
... The rest is on how to improve the creation of standards.
... The basis for a sustainable business model is the importance of the Web.
... Thank you for creating the next ten technologies for the Web. It is those technologies that create the sustainable business model of the W3C.

Jeff: We've focused over the last year on how the Web isn't just technical, but transformative.
... The important thing about the Web today, the Open Web Platform, is going to transform every single business once again.
... Industries are changing from the work we do in these rooms here. Devices, Web of Data, security… all of these companies need to be aware of that.
... The core of the success of the messaging we've put together can be evaluated in several ways. The way I prefer to evaluate it, because it's succinct, is what I call 18 in 18.
... We've added 18 new Full Members to the W3C in the last 18 Months. 25% of our Full Members are now brand new.
... There is plenty of renewed interest. Significant brand names across a wide set of industries.
... 2011 Focus on HOW we create standards
... Getting started early, finishing on time, headlights, getting the right participants/stakeholders, how we talk to one another, developer conference, and to end, is it time for us to re-evaluate W3C process?
... Getting started earlier: Community Groups. We looked at the difficulty of getting innovators to the W3C. We started with Incubator Groups, which were good, but not good enough. So, we launched Community Groups.
... CGs are free, open, and easy to get developers to W3C.
... Anyone can participate, even non-members for free.
... There's latitude in how CGs work is structured. RF and IPR are different than WGs, but can still lead to Rec.
... 30 CGs formed, 12 more proposed.
... Timeliness: Missing schedule hurts
... Charters list when recommendations come out. We usually miss the schedule.
... Good or bad?

Roger: Bad.

Jeff: On average it is bad. We have to be careful in the terminology. There's a balance between right standard late and wrong standard on time.
... After looking at lots of data, we were too late.
... Sometimes things change, and they have to be late, but that is a minority of the situations.
... The cost is that we can't get as much done, damages our reputation, etc.
... Metrics, measured the number of projects that I looked at. Looked at the dates on those projects, and what we originally targeted.
... What percentage are on schedule, and how late were we.
... Looked at 2 categories of projects: how many got to rec in the last 12 months: 16. How many are sitting there waiting to be completed: 147.
... Of the 16 we completed in the last 12 months, 9 of them we have an original date for, and of the 141 we know 91 of them.
... We finish 10-20% of things on time, roughly.
... We're talking 100 things and finishing just 10-15% of them on time. That might be okay if we were just consistently 6 weeks late.
... But we're not.
... On average they were three years later than originally chartered.
... Things we completed were almost 3 years late. On going is 27 months, but that's not less than 3 years, because it's 27 months and counting.
... Ideas for Chairs
... We have some ideas of what the team can do better, and what W3C management can do better.
... I want your input on this, that in an hour or so when Ian and Tantek talk about the breakouts that they propose one on this question.

Kai: Just wanted to point out that we already have a breakout session for that called: "Revisiting the way W3C creates specifications"

[[Session 3 in California room 1]]

Jeff: We'll leave it up to Ian and Tantek to determine if they're the same, if they are great.
... How do we make sure we're noticing it on time. We're doing work internally to the team, but presented to membership in time, to make sure the right topics are chartered in time.
... We're getting better about getting new things started quickly, WebRTC, Security and Privacy, etc.
... But we can do better.
... We're going to get a little more focus.
... We're going to look at new conferences, etc, and figure out where to do new work. Some will be incremental and get automatically get put into W3C process.
... In some cases there may be a whole breakthrough new dimension that requires new outreach and innovation.
... In that case we'll go through a new W3C "Headlights" process. Every winter we'll see if there are new big things happening.
... The hope would be then to socialize it and get it to the members at the spring AC meeting.
... Then every summer we have our internal strategy conference. There we decide resources, etc.
... Another thing we'll do for new things is that we sometimes get a situation where a WG has started, but it doesn't get traction or doesn't get adopted. Or maybe there are a handful of people in the WG that implemented it, but no traction because critical stakeholders aren't participating.
... That is something we need to manage better at W3C.
... I've asked the AB if this is something they should be doing, and they said no, we should be, so we're doing it.
... So team contacts have to assess their WGs, identify lapses, keep in touch with implementers and report weekly.
... Domain leads track that and reports to W3M quarterly or as needed.
... W3M figures out where we need help.
... Purpose of TPAC is for developers to work together.
... We've heard in the past that a full day of presentations or panels or just being talked to wasn't as meaningful as some would like. The purpose of TPAC is face to face meetings, many WGs working together. The generalization is that we're all people and need to be talking to one another.
... So, let's make TPAC a giant opportunity to get people to talk to one another, we'll see how it works out.
... I am hopeful and anxious about how it works out.
... After brief plenary, there will be breakouts, then a brief closing plenary for sharing breakout info.
... A few announcements:
... We've created our first conference to focus on the needs of developers and designers. W3Conf!
... Thanks to Doug for driving this internally at W3C and thanks to Microsoft for their generous sponsorship.
... You can still register to be in the audience or join the streaming.

Doug: And thanks AT&T, Nokia and Adobe for also sponsoring!
... There are still a few seats available. If you register with code 'tpac' you get $50 off the registration.

Jeff: That's two weeks from now.
... Next week we have the W3C Social Business Jam.
... That's where we figure out what social means to business.
... Of the items I talked about we probably spent the most time talking about Community Groups.
... We need to understand whether some of the CG practices can filter in to W3C WGs as well.
... Changing our WG process isn't something that changes on a dime, but I am interested in what changes are needed. Who should be involved? Who are the stakeholders? What are the priorities for what gets solved?
... I hope this was useful. I went a little over time, but let's take one or two questions.
... Let's go directly to the panels then.

Web and TV IG HTML 5 Proposals

Mark Vickers, Comcast

Mark Vickers

Mark Vickers

Mark: I am Mark Vickers from Comcast, we've been running an interest group on Web and TV.
... Focus has been to look at use cases and compare them to the specs.
... Most use cases were fully supported, but there were gaps.
... The members of the group, I've split into three categories: media/operator, research/government and electronics/software products.
... The group is about the media and operator folks delivering content to electronics and software groups.
... There's nothing specific in this group about TV as a specific device. There's no TV specific API.
... However there have been TV and media related gaps that have been identified, but they apply to all devices.
... Media delivery has been moving to the cloud over the public internet, rather than private networks.
... The HTML 4 world is pretty much what we deliver today.

Mark: There's nothing in the HTML 4 spec that gets you video. All video is done through plugins.
... Our world is between HTML 4 and the plugin.

<tantek> I disagree with the assertion that "There's no TV specific API" - TVs often have various hardware capabilities that are not common on other devices, e.g. multiple video/audio inputs to choose from, picture-in-picture etc. All of those could inform the design of a "TVAPI" in the broader category of WebAPIs.

Mark: In television there's a mixed world, we deliver to televisions now. Some are HTML 4, just binary, some widget like things, apps.
... We've been working towards new hope of HTML5.

<Alan> ArtB: Clarke Stevens (CableLabs) & Giuseppe Pascale (Opera)

Mark: We want to deliver content to any HTML5 Client.
... You can add subscription based, on demand, or broadcast, but this group has been focusing on where the issues are.
... The first area is about the delivery.
... In professional video we don't have just one video and one audio, but through the work of the XG people and W3C, HTML 5 has good multi-track.
... And there are many many data tracks associated too, captions, interaction, content advisories, etc.
... So it's really a bundle of things that are sent out together. That support is in there, and you'll hear Clark talk about that and what tweaks are needed.
... There's also adaptive streaming based on available bandwidth.
... Then there's content protection.
... In HTML4 this happened through plugins.
... In the open web there is DRM, which is done with plugins. Those have been approved by the studios.
... For home use there's DTCP-IP, and there are plugins for those too.
... HTML 5 supports content protection, but it's up to the browser to support it.
... Other options for content protection, something open source and open standards based, but new protection can be brought to the studios and approved over time.
... All of those are in the Media Pipeline Task Force, which has been focused on that.
... The other aspect is that there is another cloud. There's a different cloud, maybe a fog or something, the cloud of the home.
... The connected devices of the home.
... These are connected through technologies that Giuseppe will be talking about.
... Browsers haven't had access to these technologies, but could. Phone as remote, or bluetooth connections, etc.
... That is the Home Network Task Force, extensions to provide support for the standard Web.

<scribe> scribe: MichaelC

Clarke Stevens, CableLabs

Clarke Stevens

Clarke Stevens

cs: my kids watch video on the web now, not tv devices

some of it is proprietary formats

no standard holds back professional industry

industry cares about content rights

basically, if you don't have a DRM solution, you won't get the content

it's also not practical to support arbitrary number of protocols

features from mainstream TV like content advisories, authorization need to be built

accessibility solutions not standardized yet

but of course we want to go beyond the traditional model, if we're moving to the Web

e.g., multi-screen

multiple content types

platform customizations

interactivity

context awareness

(such as location, have program follow you as you move among devices)

strategic use of available bandwidth

-- requirements for the above --

Combined main and descriptive audio track

Handling in-band tracks

(program, ads, etc. all coming together)

Ability to pass parameters to server such as device capabilities

to adapt bit rate etc.

and send feedback about how things are working within those parameters, to adjust

Content protection (i.e., DRM)

planning to file issues on HTML 5 to address gaps in above requirements

need to work with HTML WG, accessibility WGs, etc. to move these requirements forward

Giuseppe Pascale, Opera

Giuseppe Pascale

Giuseppe Pascale

<inserted> http://people.opera.com/giuseppep/hntf-usecases-gaps/

<dom> Giuseppe's slides

gp: Q: What do all the devices people use have in common?

A: they all run a browser, and all run on home network

huge percentage of people watch TV and surf internet at same time

use cases for multi-screen

select video to watch on PC, then send to TV

stick everything onto one screen

use smartphone as device remote control

supplemental content to program on auxiliary devices

when we have connected devices, can meet these use cases and others

the technology is almost there

but you have to download application, it has to discover your devices and be able to communicate with them

no way for device discovery exists right now

both their existence, and available services

existing mechanisms can support communication between devices

but need to address communication between two client user agents, wrt same origin restriction

Home Network Task Force has determined the device discovery is the primary gap needs focus

could use XHR if relaxing same origin restriction

There have been a series of Web & TV workshops

today, want to explore some of the results with groups here

Friday there is a joint Web&TV and DAP meeting to start technical work

Clarke Stevens, CableLabs

cs: have implemented API from CableLabs as an applet

<dom> Discovery API demo (Member-only)

which tries to address issues raised in the presentations this morning

Roger_Cutler: think there could be security implications, is this being investigated?

Mark_Vickers: not really being explored yet, but now is the time

Ileana_Leuca: suggest meet with WebRTC

some areas of common interest in the technical solutions

Ralph_Brown: how do we instill urgency in this work?

mv: many of the issues already filed on HTML 5, as tweaks of existing features

Home Networking API is new

looking for where that would go

cs: demoing 2 UPnP servers and a DAP server

start discovery

<technical issues scupper demo>

UA should present a dialog box listing discovered devices

user indicates which ones to tell Web page exist

then can use XHR to communicate between them

use can change which devices exposed to the network at any time

<dsr> scribenick: dsr

web content interoperability

Web Content Interop Panel

Bryan Sullivan, Wilhem Joys Andersen, Soonhoo Lee, Kai Scheppe

chaired by Bryan Sullivan

Why worry about content interoperability? There a somewhat accurate perception among developers that developing for the web is hard, especially for the mobile web.

why does that perception exist/persist? What are your expectations for content interoperability? what are the top issues for W3C to focus on?

e.g. the role of the W3C Testing activity for verifying consistent behavior across browsers.

Developers want it to just work! It doesn't have to be totally easy - after all we live in the real world.

So what do you think are the key interoperability issues?

what can be addressed by more/better testing

what can be done with better specs or improved W3C processes?

what issues can be addressed by supplemental approaches, e.g. JavaScript libraries, developer guidelines etc.

Kai (Deutsche Telekom) we care about content interoperability and our designers traditionally cared about pixel for pixel accuracy.

It was a battle to convince them that that isn't the big issue.

Dealing with variations across browser versions would cost about one million euros, something no one will pay for,

We therefore focus on a small number of browsers and versions.

Bryan: those costs get in the way of other work, e.g. on improved accessibility.

Soonho Lee, SK Telecom, mentions the ability for developers to sell apps in app stores.

customers will be upset if the applications that they have paid for turn out not to work properly.

Bryan: the economic model for native and web apps is different.

Soonhoo Lee: we need to reduce the costs of testing, better interoperability will help that.

Claudio Venezia (Telecom Italia) if there is a core set of interoperable features that would help. This is better than large but poorly implemented specs.

Bryan: if the specs take too long to finish, interop testing can't happen in a timely way.

<Josh_Soref> [ Laughter ]

Wilhem Joys Andersen (Opera) In the late 90s, we taught web developers that specs cannot be trusted. We taught them that each browser required different instructions to perform as intended. In part because browser vendors implemented their own thing with no regards to the standardization process and in part because the standards bodies failed to write specs and standards for the technologies that were actually used.
We've spent the past 15 years cleaning up the mess we created. Browser vendors have had to make the web more interoperable one broken site at the time. It's been expensive. It's been painful.
But I believe both the W3C and browser vendors have learned from all these mistakes.
We have mostly, but not completely, stopped writing fiction. A spec without a good test suite is fiction. A spec without implementations is fiction. A spec without content is fiction.
We have learned that no plan ever survives contact with the enemy, and no spec ever comes unharmed from implementation, testing and content production. Specs are not delivered on stone tablets from the heavens. They are developed through a chaotic, frustrating, iterative process. The spec influences implementations and content, but is also influenced by them. Even after REC.
We have learned that the Web is fragile. It's a wonder this thing even works. If you fix a browser bug on Facebook, you are likely to break GMail. To ensure that implemenatations remain interoperable, we need the official W3C test suites to be good and extensive. HTML4 had a few hundred test cases. Based on my experience through seven years of testing browsers, I believe we need no less than one million test cases to properly test the Web platform.

Ann Bassetti (Boeing) we have the same problem in big companies and our developers struggle to understand how to address the interop issues for the Web.

Kai (DT) W3C's mission to lead the Web to its full potential - this is being hurt by the interop problems

Roger: W3C seems fixated on the browser, there is a lot more to consider

Ann: we also are interested in a wider range of devices than desktop and mobile.

<brutzman> not hearing an accessibility story... are they thinking interoperability first, maybe accessibility later?

Noah (TAG) this community puts a huge amount of effort into being liberal in what the browsers accept,

but perhaps less on other aspects.

Mary Brady (NIST) testing is hard! We would be happy to share our past experiences with you.

<ArtB> ACTION: Mary Brady get NIST to submit the 1,000,000 test cases Wilhelm said we will need for HTML5 [recorded in http://www.w3.org/2011/11/02-tpac-minutes.html#action01]

Charles (Opera) there is a feeling that real men write HTML in a text editor, but good authoring tools are a really important part of the puzzle.

Bryan: the integration of Web APIs in to good authoring tools is really valuable.

Philippe (W3C) millions of dollars spent on testing for big websites. If at W3C we don't put the required effort into testing for HTML5 we will be suffering for a long time into the future.

<koalie> Web Testing interest group

Kai: the answer doesn't lie in perfect specifications that take a long time to finish, better to crowd source smaller specs.

Daniel Glazman (Disruptive Innovations) we lack editing coordination across W3C specs

Bryan thanks everyone for their comments.

[ Applause ]

Agenda building

Agenda building

Around the grid, agenda building

<koalie> BarCamp/BreakOut Session Grid

Ian explains how the rest of the day will work

Ian thanks the members of the TPAC plenary program committee

Tantek explains about bar camps. Every session today have been written up on the wiki, see http://www.w3.org/wiki/TPAC2011/PlenaryBreakouts

Tantek asks people to introduce themselves to the people seated behind them.

(lots of noise as everyone starts talking)

Tantek calls us all to order....

Ian: we are going to spend the next 20 minutes to work on scheduling the break out sessions.

<dom> Schedule of the break out sessions

Each session is about an hour long and the first will run until Lunch

<Josh_Soref> [ Crickets ]

There will then be 3 more sessions and we will then come back for a show and tell report back.

<dbaron> Should there be a time limit for the sharing presentations?

<dom> Session ideas on W3C wiki

Tantek: some months back we decided to invite topic proposals on a wiki. We preselected a small set of them. You don't have to go to them.

The actual schedule is picked today.

You will be invited to come to the front and write your breakout proposal on the green sheets: please write 3 things - the subject of the session, the name of the discussion leader(s) and the hashtag for the IRC channel and for twitter if you plan to use that as well.

You don't need to take a full hour slot if you don't need that.

There is a record number of people at this TPAC, so space is at a premium. This makes it a good idea to merge closely related sessions.

There are also some overflow areas for smaller sessions.

Ian encourages people to come to the summary session at the end of the day and for each breakout to assign scribes to take notes on the IRC channel.

Tantek - be sure that the scribe links to the grid on the wiki.

Tantek invites people to come forward now to post the proposals on the whiteboard.

<Josh_Soref> [ Cricket ]

<Josh_Soref> Anne: We want to reserve a time slot

<koalie> [people break to break-out rooms]

<jeff_> Where is the grid posted?

<amy> grid: http://www.w3.org/wiki/TPAC2011#Session_Grid

<shawn> pointer to page with sessions?

<hober> shawn: http://www.w3.org/wiki/TPAC2011#Session_Grid

<shawn> thanks!

<hober> np

Sharing and Advocacy Plenary

<jeanne> scribe: jeanne

Laura: Building Cloud Applications. Started on cloud security, but the group was more interested in standards around cloud applications. How can we leverage existing standards to apply to cloud

<koalie> TPAC2011/Building Apps in the Cloud

Laura: what can W3C do in addressing issues now and going forward.

Doug: Developer documentation and education: We talked about the need for better documentation
... to help people create content.
... discontinuity between developer documentation and specifications.
... HTML5 has a mode that hides and just shows developer documentation

<brutzman> can someone please link "Declarative 3D" on the meeting matrix at 1430 in room B to http://www.w3.org/community/declarative3d/

<dom> [the W3C cheatsheet extracts data from specs to build a dictionary of markup elements, css properties, etc, cf http://dev.w3.org/2009/cheatsheet/doc/developer.html ]

Doug: how summaries in working groups would be very useful. "This week in X working group" so people would understand issues in specs
... reaching out to international contacts to increase translations

<dom> http://w3c-test.org

plh: Testing in Web Browsers: update on status on the current work. There is framework on a mercurial server w3c-test.org. There are a few working groups using it.
... showing testharness.js

<dom> http://w3c-test.org/html/tests/resources/testharness.js

Bryan: API Design Approach: Data minimization for privacy protection

Brian: make sure API is targeted to the audience
... the design should follow a progressive path

<RalphS> [Bryan reviews -> http://www.w3.org/wiki/TPAC2011/API_Design_Approaches_and_the_Rationales_for_Them Results ]

[summary on the web page]

<koalie> minutes of Global Participation Breakout

Ann: Encourage more global participation in w3c and web
... we think people should speak more slowly with simpler English

[applause]

Ann: part is the W3C does very technical work. Timezones. Email thread that is fast-moving, may have a decision before participants in Asia even wake up. Unfair.
... quiet voiced people have a hard time getting attention on a call

<brutzman> Japanese gamer in the HTML5+Games breakout described "globish" = Global English 8)

Mani: Challenge of Identity
... enabling good user experience is to insure that the user knows who they are talking to. Two-way communication with privacy safeguards.
... applicable models like OpenID for authentication.
... we don't trust the service provider, the email ID is the most ubiquitous identifier.

Alan: Social Business Jam: we talked about how the JAM will work, it is international, there will be 6 topics and anyone can participate 24/7 for three days

Tab: Agile Standardization within the W3C Process

<RalphS> Agile Standardization Summary

Tab: Modularize - interconnects can kill you. FOcus on a bedrock spec as fast and small as possible.
... then build modules on top of that that only refer to the bedrock spec and limit the interdependencies between them
... stages: Exploratory, Stable and Recommended
... blog post from fantasai

Hadley: Open Gov Data
... governments decided that anonymous data should be available for others to use. But it is not standardized
... we discussed transfer between formats, documentation of the data
... proprietary formats and licensing problems
... linking to new formats of the data created by a 3rd party
... need for new vocabularies - different language used by govs and developers

Paul: HTML WG Decision Policy
... explained the history, why WG needed more than the W3C process
... a surprise that the decision policy could evolve without a charter change

Paul: concern that the editor handled the bugs and then the bugs can be escalated
... informative session

Jeff: Fixing Schedule Delays
... started with Jeff's ideas, which were thrown out and started with a blank slate.
... 23 ideas, narrowed down to 16
... create a separate role of test editor for every document
... create concept of a living working group that will last forever

<koalie> TPAC2011/Fixing schedule delays

Jeff: implementors need to make commitments earlier in the process

<ArtB> Minutes from scheduling delays: http://www.w3.org/2011/11/02-schedule-minutes.html

Johannes - Declarative 3D. Only Community Group breakout session.

Johannes: put 3D as a declarative language in the DOM.
... how the system evolved over the last few years and what could be the outcome of the community group
... the main plan is to get the requirements and use cases
... then showed demos

Dan: Right to Link
... copyright and publishing data on the web
... inform opinion of people in regulatory and policy arena, on what it means to host web sites. So that policy makers understand web publishing.
... definitions of terms
... guidance for terms of service
... fair use, reuse of data, content, and publishing
... contribute thought to Dan or TAG mailing list

Kim: Explosion of Input Methods

<koalie> TPAC2011/Adjusting to explosion of input methods

Kim: there were the keyboard and the mouse. Now there are 5 input methods - touch, speech and gesture
... less hacking, most systematic approach, giving users more choice
... we need to talk across groups. Intentional Events, EMMA

HTML5 and Games: Game programming HTML5 for Kids. Shows cartoon showing W3C

<koalie> TPAC2011/HTML5 and Games

Raman: Evolving web platform Web APIs and accessibility

<koalie> minutes of breakout on Web APIs and Accessibility

Raman: web platform is being to include very rich interaction APIs, speech. Web apps do all the work instead of letting the UI appear elsewhere.

<brutzman> (late entry) the Declarative 3D breakout session description is at http://www.w3.org/wiki/TPAC2011/SessionIdeas#Declarative_3D_Graphics_on_the_Web

Raman: services on the web - google maps, twitter - allow the developer to build light weight web apps with different UIs based on the service.
... now we have the ability to deliver apps that meet the needs of every individual user.

<dom> re HTML Speech in the browser, cf http://lists.w3.org/Archives/Public/public-xg-htmlspeech/2011Oct/att-0064/speechwepapi_1_.html

Kai: Revisiting how W3C creates standards

Kai: we had so many input points we decided to brainstorm

<dom> [A community Group has been created in result of the session Kai is presenting: http://www.w3.org/community/w3process/ ]

Kai: specs are too large, too complex, and too long

<RalphS> [Kai summarizes -> http://www.w3.org/wiki/TPAC2011/Revisiting_how_W3C_creates_standards Revisiting how W3C creates standards ]

Kai: the process document doesn't match the development
... hard to fix charter
... drafts are continuously outdated
... hard to determine what the changes are
... specs are only looked at during Last Call

Kevin: HTML5 AV Club session - as opposed to putting more DRM in HTML5

<koalie> minutes of breakout on HTML5 AV Club

Kevin: missing - the ability to create a playlist
... adaptive streaming discussion
... the difference between a short news program vs. a streaming movie.

Kevin: the ability to save media

Paul: jQuery and js developers want from web standards:
... feature detection

<koalie> minutes of breakout on What jquery and js developers want from web stds.

Paul: detecting what the best features are of each browser
... documenting
... use cases of benefits of the spec
... community for authors to interact with the standards community and getting their feedback listened to
... standards work email lists are scary, and there need to be ways to attract their input.
... a safe place to propose ideas

IanJ: this is the second community group, and it may be an option for your group as a next step

Alex: Linked Data

<koalie> TPAC2011/Linked data

Alex: how can we communicate to people. We don't have the right resources.
... what is the difference with Open Linked Data.
... we put more emphasize on enterprise and linked enterprise linked data especially data integration
... there will be a workshop in December
... TimBL's 5 stars of Linked Data
... discussed forming a community group.
... also about studying patterns

<koalie> TPAC2011/Web-based Digital Signage

Shinji: Web-based Digital Signage
... features and functions, use cases in Japan

Robin: Improving W3C Publications Ecosystem

<koalie> TPAC2011/W3C Publications Ecosystem

Robin: hard in terms of producing specs that match existing specifications
... style and boilerplate

<dom> Follow up conversations on publication ecosystem will happen on http://lists.w3.org/Archives/Public/spec-prod/

Robin: some css hackers are going to work on that and develop some icons to improve understanding of status
... html5 will soon be able to be used in specs

<anne> yay

Robin: rich UI in specs - linking to test suites, comment on spec, etc
... tooling - common output, inserting references
... wikis to create specs, rich semantics
... W3C needs a Managing Editor to be a go-to

Debbie: Demos of W3C technologies
... CSS regions, layouts
... Open Stream, medical history on tablet or handheld medical history tool that can dictate or add videos
... InkML - interoperability of Powerpoint, HTML5 and chat

Ted: HTML5 and Games
... new community group of people who were interested in games on top of HTML5.

<koalie> TPAC2011/HTML5 and Games

Ted: report from the HTML.next for Games
... need better audio support for games

Ted: developers don't know what to use, and browsers don't know what to use, so we are trying to improve the communication
... make real requirements to go to working groups

minutes of breakout on Content Protection

Mark: Content Protection
... it is clearly not an objective to standardize in HTML5, but there are capabilities that need to be exposed so that commercial services can be supported.
... if you try to play back protected content, the key exchange can be mediated at the service layer
... need for more descriptive error codes when things go wrong
... this can be reported back to the customer service of commercial companies.
... there will be more discussion about this.

IanJ: please give feedback
... thanks to the program committee

<koalie> Feedback survey

<fukuno> W3C girl here -> http://fukuno.com/w3c_girl.html

Wrap up

TimBL

TimBL Wrap-Up

<koalie> scribenick: amy

TimBL: I got mutiny/desertion. Jeff pointed out that he'd been jumping up and down all the time and I had hoped he hadn't noticed I'd be sitting there having fun
... he's dropped out of the wrapup. just me
... I wanted to say thank you. i want to say my own thank you to Ian and Tantek for the unconference thing (clapping)

TimBL: I've done a few unconferences. this one, I think, though I may be biased. i feel people were really engaged
... I didn't go to every room but the ones I felt the vibe from, people were there w/ acute interest, acute need to communicate or resolve.
... from my point of view, from my own need to get connected into things, it passed the connected sessions test. the urge to be in more than you really can be. i regret missing some
... that made me feel to a certain extent that my thirst had been quenched but my appetite whetted to learn more.
... this is a massive task, building the next web. there is a huge amount of work. there is a huge amount of coordination
... one of the interesting things it to take this and ask to what extent W3C can be an un-organization
... if we can do this f2f, to what extent can we do this to W3C
... W3C has always been a group that decides it's own process, makes it's own agenda.
... i have a feeling that lots of groups were talking about architecture, modularity, etc. ".which way should we take things?"
... different groups are working out what it is we need in the future. can we, you, us, together, when we go home, when we go back online, can we work in a distributed fashion to address communal things.
... communal things like consistent architecture. testing. i love that everyone is talking about testing. that's great. it's great that we're taking testing seriously
... doing architectural design, that's more of a test of the un-organization

TimBL: another issue, managing process. i liked the group meeting by the bar from which I was politely excluded. they re-engineered the W3C, reconstructed

TimBL: different places where people are looking at publication, HTML5 group
... difficult things but very important

Crowd Shot

TPAC reception

TimBL: I'd like to leave w/ you to be communal. think about this being an unconference. think that if you don't do it maybe someone will.
... if you organize other people's lives/architecture. you have to be respectful
... i hope people will find this a good time for talking not to people you think are right. but also to those you think are wrong. go find them. find out where they're coming form.
... you have alcohol at your disposal.
... this happened because of tantek and Ian. Jeff also had a lot to do w/ this. Thank you Jeff for running the org
... mainly it is the admin team.. all the stuff we never heard about
... i leave you w/ this challenge. we've had one day of unconference. think about your role in an unorganization. think about the process. let's make it something w/ works on mutual respect. something peer-to-peer.
... everywhere we've made a rule (sometimes because we've done something wrong). we put constraints and boilerplate. let's review them. love the idea of replacing w/ icons. let's think about new ones. we'll get them wrong but on balance. we should be constantly be modifiying our thoughts about architecture
... last people that I thanked, front half of the room stand up!! stand up (I mean it!)
... everyone stand up, up to Coralie, stand up, turn around everyone else stand up, as you're able
... everybody clap (clapping). well done. thank you for coming

TimBL: until next time
... reception, 6pm end of the hall


Minutes coordinated by Coralie Mercier <coralie@w3.org>
Photo credit: Kaz Ashimura
$Date: 2011/11/10 22:41:48 $