Social Web Incubator Group Teleconference -- 28 Oct 2009

<trackbot> Date: 28 October 2009

Convene SWXG WG meeting of 2009-10-28T14:00-16:00GMT

<mischat> zakim, +0162272aabb is me

<mischat> i have done it 3 times :(

<DKA> Scribe: Dan

<DKA> ScribeNick: DKA

Harry: any comments on the agenda?

<hhalpin> PROPOSED: to approve SWXG WG Weekly -- 21st October 2009 as a true record

<hhalpin> http://www.w3.org/2009/10/21-swxg-minutes.html

<hhalpin> RESOLVED: SWXG WG Weekly -- 21st October 2009 are a true record

<hhalpin> PROPOSED: to meet again Wed. November 11th (Note that we are skiping the meeting of Nov. 4th due to conflict with TPAC)

Harry: next meeting will be Nov 11 - since we will skip next week due to TPAC.

<hhalpin> RESOLVED: Wed. November 11th

General Organization

Dan: Should we have dial-in facilities at TPAC?

Harry: we need to reserve this now if we want it...

<tinkster1> Probably not

Harry: would anyone dial in if they are off-site?

<hhalpin> If anyone wants to, they can.

Harry: I will send an email out to the list - if we will get more than 3 people then I will request it.
... I will include the agenda in the mail I send.
... Any brief updates on TPAC or social web camp?

<bblfish> http://barcamp.org/SocialWebCamp-Santa-Clara

<bblfish> There are 68 people signed up

Dan: update on the camp - the last I saw there were 60 people singed up -
... http://barcamp.org/SocialWebCamp-Santa-Clara
... Can people please send that around / facebook / twitter / identi.ca / whatever it... use sky-writing, etc...

<bblfish> Sun is paying $10 for lunch

<hhalpin> +1

<CaptSolo> Zakim: aadd is me

<hhalpin> David Morin from Facebook.

Harry: Other good news on TPAC - we have a number of observers coming in...

<mischat> there is an echo

Harry: Any updates on actions?

<hhalpin> ACTION: [CONTINUES] tinkster to summarize Evan's talk [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action01]

<tinkster1> I've posted that to the list already.

<hhalpin> ACTION: [CONTINUES] DKA to summarize OSLO and geoLocation conversation in order to spread knowledge of these efforts among W3C members. [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action02]

<rreck> tinkster did you get my email?

<mischat> hehe, i have another couple of actions

Harry: Giant congratulations to mischat for finishing his action.

<hhalpin> ACTION: [CONTINUES] Mischa to describe/implement a report of terms and conditions, and how they change between now and the end of the XG. [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action03]

<tinkster1> http://lists.w3.org/Archives/Public/public-xg-socialweb/2009Oct/0051.html

<tinkster1> Yes.

<bblfish> there is an echo

<rreck> i think is almost done

<tinkster1> I think it was changed to DONE last week.

<rreck> mine

<mischat> harry is echoing

<bblfish> perhaps everyone is

<rreck> echo is based on noise canceling

<tinkster1> everyone seems to be.

<mischat> yay better

<hhalpin> ACTION: [CONTINUES] Adam to write up Matt Lee's talk [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action04]

<hhalpin> ACTION: [CONTINUES] mtuffied to put up wiki page about social networks deploying these technologies. [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action05]

<hhalpin> ACTION:oshani to write up David Recordon and Luke Shepart talk. [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action06]

Invited Guest Invitations

Harry: Ted will be talking about applying social tech to w3.org...

<hhalpin> ACTION: [CONTINUES] cperey to book global lockbox as an invited speaker [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action07]

<mischat> i have no joy with BBC persia

<hhalpin> ACTION: [CONTINUES] mtiffiel to invite BBC Persia people to talk about their use of social media [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action08]

<mischat> nope will try calling them again

<mischat> will hassle

<mischat> for another week

<mischat> :)

<jsmarr> hi i'm on the call

<hhalpin> ACTION: [CONTINUES] petef to look into activitystreams invite, maybe Chris Messina [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action09]

User Stories

Harry: we can talk user stories and final report next week [at TPAC].

<hhalpin> Will talk about this next week at TPAC.

<hhalpin> ACTION: [CONTINUES] adam to write up the boeing use case for enterprise social networks [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action10]

<hhalpin> ACTION: [CONTINUES] rreck to flesh out anonymous usecase connecting to multiple identies and null provenance [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action11]

<hhalpin> ACTION: [CONTINUES] oshani to reframe the geolocation/intent/portability [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action12]

<hhalpin> ACTION: [CONTINUES] bblfish to relabel data protection use case to be about controlled access and takedown to data "about" you [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action13]

<rreck> mine is almost done

<hhalpin> ACTION: [CONTINUES] bblfish to merge Family and Group access usecases [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action14]

<rreck> i promise

<hhalpin> ACTION: [CONTINUES] hhalpin explain to henry and oshani doc editing process for usecases [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action15]

Final Report

<bblfish> yes, I have not had a lot of time to do much, on my use cases as I have been quite busy organising http://barcamp.org/SocialWebCamp-Santa-Clara

# Invited Speaker: Joseph Smarr from Plaxo and Portable Contacts

<hhalpin> http://josephsmarr.com/

<hhalpin> http://www.plaxo.com/

<hhalpin> http://portablecontacts.net/

Joseph Smarr: I'm CTO at Plaxo - Plaxo's about contact management, keeping together sources of contact data...

<hhalpin> http://portablecontacts.net/draft-spec.html

Joseph: Lately I've been involved with the so-called open stack. This is also strategic for Plaxo.
... A lot of these technologies are making progress - but everything on the Web wants to add a social component. In order to get that people data
... reconnecting to those people you know is daunting. Most sites are scraping your data using your username/password.
... until now the tools weren't available to do anything else. Every single site has built their own contacts protocol / db / etc...

<hhalpin> good point re vendor-specific APIs over scraping

Joseph: genesis of portable contacts came out of [trying to solve this problem]. Because even APIs when they are proprietary are not solving the problem.

<rreck> ha!

Joseph: someone else building contact API should be able to use them... With portable contacts there should be a network effect -
... so we get away from this vendor-by-vendor integration. This is more how the Web should work.
... portablecontacts.net

<rreck> simple sounds like a good idea

Joseph: getting adoption is the hardest thing - the best way to get adoption is to make the protocol simple ... minimal tool-chain complexity...
... it's not just having a standard schema for contacts, but specifying how you authenticate, etc... end-to-end. But each piece is "not invented" as possible.
... Schema is taken from v-card, etc...
... the discovery protocol is XRD. The auth protocols are http basic or oauth. Simple REST-style [API].

<hhalpin> :) pausing for breath!

Joseph: that's basically where things are at. It's a draft spec for a year. Plaxo, Google, MS, [others?] have implemented...

<hhalpin> Myspace has implemented, and part of OpenSocial.

Joseph: have also got together with the open social folks - they have specfified REST-ful protocols for contacts. They did something similar. We worked with them to make open social a subset of portable contacts.

Adam: have you given thought how it might have value to the enterprise?

<hhalpin> Enterprise deployment of PortableContacts...

Joseph: portable contacts useful as itself as a useful protocol - so if you're building applications that share contact info [even within the enterprise] then I suggest using portable contacts format.
... if you can use portable contacts then you only have to build it once if you want to integrate with other social networks, etc...

<Zakim> danbri, you wanted to ask how far schema extensions can be supported in the protocol

joseph: the other thing we do is helping to educate market...

<rreck> louder please

Danbri: How far can the schema be extended without [breaking things]?

<danbri> q: how far can schema be extended beforeyou go beyond what protocol can understand? what's the plan for navigating that tension?

<mischat> :)

Joseph: we like the html model of extensibility - if you want to extend then you can and we'll ignore it if we don't understand it.

<rreck> +q

Joseph: within each field there is a type value and each in each type values there are canonically specified values but you can use your own as well... the auth protocols are also extensible.

<hhalpin> So we have examples of those extra fields.

<hhalpin> Wondering if there's a wiki where these can be listed...

Joseph: it's intentionally quiet around namespacing, extension mechanisms, etc... entire spec is intended to be extensible.
... Also - there's already been a lot of talk of using the schema in other contexts - e.g. webfinger.
... some work on an XML namespace is beginning. We could use some help and guidance on that.

<danbri> http://groups.google.com/group/portablecontacts/browse_thread/thread/6954a050d65485bf/ced1b56aaa246a85?lnk=gst&q=brickley#ced1b56aaa246a85

<danbri> some discussion of xml ns there

<danbri> grddl etc

Harry: On that topic - it' would be interesting to see more alignment - between FOAF and portable contacts.

Dan: What's the IPR policy of portable contacts, etc...

Joseph: We are looking at open web foundation or something... We're not going to make a separate foundation. Most of the people who have contributed to the spec understand that it will go in that direction.

Harry: [relating the vcard schema effort...]

<tinkster1> http://microformats.org/wiki/jcard

Harry: [circular statements that have broken my fingers]

<rreck> woman's voice cross talk

Joseph: Apple is spearheading a card-dav effort - which is not using the XML schema.

<danbri> in http://portablecontacts.net/draft-spec.html#schema i'd love to see a distinction between PC core, and the opensocial stuff which seems less firm

Joseph: there wasn't much in there [vcard xml schema] that was controversial.
... it would be good [to align these]
... if you are interested then please say so on the [ietf?] mailing list.

<hhalpin> IETF VCard 4.0 list.

Joseph: Needs to be linked up to JSON format. JSON can be serializable in modern languages like PHP, etc... so no need for parsing.

Harry: it appears that portable contacts is a proper superset of vcard3.

<danbri> (also re I18N of names ...)

<hhalpin> 1-to-1 mapping above vCard 3.0 and modernized some spellings

<hhalpin> :)

Joseph: we modernized some of the spellings - adr to address. Didn't need to be as terse as older standards...

<danbri> http://wiki.portablecontacts.net/

<hhalpin> ACTION: hhalpin to check out FOAF and PortableContacts mapping. [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action16]

<trackbot> Created ACTION-102 - Check out FOAF and PortableContacts mapping. [on Harry Halpin - due 2009-11-04].

Joseph: there is a converted between portable contacts and vcard...

<danbri> http://rishida.net/blog/?p=100 re names

Harry: re additions to vcard 3 spec like tags and photos - this was taken from empirical data about what people were using, right?

Joseph: Yes.

Harry: does Plaxo keep a record of what sites support tagging, etc...

<danbri> +1 on both full field or broken out field or both

<danbri> if both, there is assumption that one is derrived from the other

Joseph: We used 2 sources for info - all the address book APIs. We used a "composite" field (for e.g. address) where you can store both a composite and broken down version.

<hhalpin> chris messina has a blog post has a big spreadsheet on contacts APIs.

<hhalpin> referred to OpenSocial (i.e. Kevin Marks) did the field study on social networking data (top 50)

<hhalpin> I remember looking for a URI for that spreadsheet!!

<hhalpin> Anyone have it to ping into IRC

<hhalpin> I saw it once..thought it was a Google Doc.

<DKA1> ScribeNick: DKA1

Ron: it would be useful to have uni-directional relationships.

<danbri> "but relationships MUST have been bi-directionally confirmed" is interesting

<danbri> assumes a b/g db

Joseph: our goal was not to model the social graph, but we wanted to express uni- and bi-directional relathionships.
... we have tags and relationships. That's meant to be used for uni-directional relationships. In Plaxo when you connect to someone you can connect as family, friends or business.

<rreck> i totally get what you are saying thanks

Joseph: when you have bi-directional relationship then you have a special relationship.

<Zakim> danbri, you wanted to ask about a PC core (that FOAF might adopt for base addressbook; then diverge on favourite-movie or trust vocab / group descriptions detail)

Danbri: on bi-directional confirmation. Does this means there is an assumption of a backing database?

<hhalpin> hmmm...just starting thinking of XFN/hCard mappings as well....

Joseph: the idea is that the semantics are relative to the provider of the contacts information. If you want to infer things like connection or relationship then you are trusting the site...

<rreck> great observation

Joseph: it doesn't [formalize] the provenance.
... if you think there's a compelling use then go build it and jump on the mailing list...

Danbri: do you plan to make any distinction of the core of portable contacts and the stuff that's coming in from opensocial?
... your schema has some things like "looking for" and some wooly things from opensocial.

Joseph: they're not formally separated but they are called out distinctly. I think the world is changing pretty rapidly [as regards what is "core" contact data].

<hhalpin> My guess is we are going to run over 10-15 minutes.

<hhalpin> Is this OK with people?

Joseph: on Facebook you care about a phone number but you also care about IM addresses another social information that people have put on [their profiles].

<Zakim> caribou, you wanted to ask about policies attached to data

<rreck> +1 run over

Danbri: how much naming stuff do you want to do? Describing family names properly for every culture / country....
... nobody's done it yet - is portable contacts going to do this?

Joseph: We try to be internationally savvy. If you have feedback about what else we could do please input it [on our mailing list]. In some cultures people have multiple names. But adding that complexity for everyone...

<danbri> when we write our final report - maybe we should suggest some joint pc/w3c effort to evaluate the i18n aspects of naming and come up with a 'good enough for everyone' strategy?

<hhalpin> +1 danbri

Caribou: I understand it's nice to transfer data from one service to another - but if I fine-tune my privacy data in one site and then the data is transferred to another site then the policies are not transferred withthe data.

<danbri> harry, can you bring that up, you've got better audio....

Joseph: Yes - the protocol is designed to only transfer the data, not the privacy metadata.

<hhalpin> ok...will bring this up.

<hhalpin> That's a place where the W3C is doing some work I think

Joseph: in a fully-interoperable soclai web this would be handled. Needs to be explored further...

<hhalpin> Carine - do you have a URI?

<hhalpin> for policy language stuff?

Joseph: There's a portable contacts google group - open to everyone.

<rreck> privacy seems against social networking business models

<hhalpin> http://www.w3.org/2009/policy-ws/cfp.html

<hhalpin> that's the w3c workshop on access.

<hhalpin> OpenID uses URIs.

<danbri> well, back to final report? i'd like us to agree that this base core (names stuff, addresses ... but not turnOns/turnOffs) deserves to be done once, well, in a way that works worldwide

Henry: [the open stack] always seems a little more complicated then it should be. Seems to be "increasing the powers of silos". Seems like the simple way to identify people would be by URL.

<caribou> hhalpin, the W3C PLING is looking at sticky policies

<hhalpin> http://www.w3.org/Policy/pling/wiki/Main_Page

<danbri> ... and PortableContacts imho is the most credible recent effort to bring together addressbook and social network field sets; I hope w3c can add an 18n perspective and assist in getting input from rest of planet

Henry: if you look at open social - the way people are identified the first four characters are split off - forcing all URLs to be URNs... It seems to force complexity.

<hhalpin> Carine - would it be worthwhile to have PLING do a presentation after the Nov. workshop?

<hhalpin> We are booked most of Nov, but Dec. is free for invited speakers.

<hhalpin> I'd like to finish up with Invited speakers by Jan so we can move to final report mode...

Joseph: to clarify - there is a unique identifier ID field for each contact. It's been a URN by convention but it's meant to be a local stable identifier. But in addition the contacts can have emails URLs, etc...

<hhalpin> I was not clear how mature the policy language work was.

Joseph: rather than it just being a URL you can explicitly break out the domain, username, etc...
... so flickr, twitter, etc... can be broken out.... to strongly identify a user across silos.
... and the fact that there is a format to use to export the data should be helping to move away from silos and towards interop.

Harry: The task fo the SWXG - we're interested in seeing what areas the W3C could play a role [or not] in the wider social web ecosystem. You could imagine something very light-weight... For example, reviewing internationalization (Richard Ishida).
... having some sort of relationship between W3C and OpenWeb Foundation.
... we could recommend that the W3C look into standardizing things. Things like portable contacts, they are already standards, but could W3C have a role here?
... any opinions?

<hhalpin> we'll have eran on I think in 2 weeks.

<hhalpin> or 3

<hhalpin> expertise around internationalization and with others standards

Joseph: I'm a big believer in standards and communities. Both as a structure for collaboration and to add gravitas. I think those are valuable suggestions. Last thing we want is a set of competing standards bodies.
... diversity is good.
... all parties should feel like their voices are heard...

Harry: One issue facing w3c is the membership structure. This incubator group is open - more or less anyone can join - but that's not normal for w3c. Is that a barrier [to participation
... ] in w3c?
... we have OWF on the phone in 2 or three weeks and we are trying to figure out what they are doing with process...

Joseph: Portable contacts has happened very organically. Mix of big companies, small companies, independents, etc... I understand why some of these bodies have rules about charter and IPR upfront, but nature of some of these efforts is light and bottom-up...
... [path to standardization] could move from light-weight process to more formalized standards body...
... to the extent that more process is necessary then absolutely.

<danbri> ok here's a quick braindump re FOAF and Portable Contacts - http://wiki.foaf-project.org/w/DanBri/PortableContactsAndFOAFLinkage

<danbri> ... can take that to the list for discussion cc Joseph

<hhalpin> 1 year and a half, pretty stable around 2009...

<hhalpin> have had as updates.

<hhalpin> or on the mailing list.

<hhalpin> ACTION: hhalpin to ping issue richard ishida to join portable contacts list [recorded in http://www.w3.org/2009/10/28-swxg-minutes.html#action17]

<trackbot> Created ACTION-103 - Ping issue richard ishida to join portable contacts list [on Harry Halpin - due 2009-11-04].

<hhalpin> yes

<mischat> bye all

<Adam> thanks joseph1

<bblfish> thanks joseph1

<danbri> If designing a form or database that will accept names from people with a variety of backgrounds, you should ask yourself whether you really need to have separate fields for given name and family name.

<danbri> This will depend on what you need to do with the data, but obviously it will be simpler to just use the full name as the user provides it, where possible.""""

<danbri> from http://rishida.net/blog/?p=100

<hhalpin> thanks alot jsmarr!

<hhalpin> Meeting adjourned

<jsmarr> thanks guys

<jsmarr> btw, is this irc transcript available/archived on the web at all/

<danbri> jsmarr, when we do a call on widget apis for social i'll bug you to join us again ... interesting links with mobile web there...

<hhalpin> We'll ping Ted Guild over looking at PortableContacts for w3.org when his time frees up.

<jsmarr> ?

<jsmarr> e.g. that i could link to for others to see the transcript

<hhalpin> http://www.w3.org/2005/Incubator/socialweb/

<hhalpin> trackbot, end meeting

- DRAFT -

Social Web Incubator Group Teleconference

28 Oct 2009

Attendees

Contents