W3C

Social Web Incubator Group Teleconference

27 May 2009

See also: IRC log

Attendees

Present
jsalvachua, cperey, rreck, libby, adam, caribou, hhalpin, tpa, petef, melvster, mischat, tinkster, karl, lubya, dom, AlexPassant, timbl, norm, pmika, ivan, Hakan, uldis
Regrets
Yuk, DanBri, DKA, Milan, Fabien, Raphael, AlexK, Renato
Chair
hhalpin
Scribe
Mischa

Contents


Convene SWXG WG meeting of 2009-05-27T13:00-15:00Z

<hhalpin> PROPOSED: to approve SWXG WG Weekly -- 20 May 2009 as a true record

<hhalpin> RESOLVED: approved SWXG WG Weekly -- 20 May 2009 as a true record

<hhalpin> PROPOSED: to meet again Wed, 3 June. scribe volunteer?

<hhalpin> Scribe?

<adam> i can do june 10th

+1

<hhalpin> OK, let's have adam provisionally as scribe for next meeting.

<hhalpin> RESOLVED: move to a scribing list

General Organization

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

apassant is a SPARQL liasion

in need of other ones ?

<hhalpin> Do we have any volunteers?

<rreck> are the groups that need liasons listed on the wiki?

<hhalpin> Yes.

<tinkster> I've put myself down for Microformats.

are in need of liasons

<jsalvachua> i may try to interface with dataportability.org

<hhalpin> Could you add yourselves to the wiki?

jsalvachua: to be liaison with dataportability.org

toby to be liason with the microformats community

<claudio> +039011228aahh is claudio

<hhalpin> C'mon no volunteers :(

<petef> I volunteered

<petef> for Social Network Portability Group List

<jsalvachua> i may help with other groups, with the vcard ietf group

renanto is on the wiki as the liason with "Policy Language Interest Group"

<AlexPassant> Uldis Bojars or John Breslin can be SIOC liaisons

<AlexPassant> her's uldis

<hhalpin> OK - could you check on them.

<hhalpin> ACTION: AlexPassant to see about SIOC liason [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action01]

<trackbot> Created ACTION-15 - See about SIOC liason [on Alexandre Passant - due 2009-06-03].

<uldis> sounds good

AlexPassant to check to see if we can find a SIOC liaison

<petef> I have added myself to wiki as volunteer liason for data portability, diso and social network portability

<hhalpin> ACTION: [DONE] danbri sketch a 5 line template for interaction with other groups (cf InvitedExperts, DiscussionTopics) [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action02]

<hhalpin> ACTION: [DONE] karl to produce a template for TF deliverables. [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action03]

<hhalpin> Karl - do you wish to explain your template?

Karl put together a template for the user stories

karl has put up 2 templates

<rreck> are user stories aka use cases?

so user stories seemed too long

<rreck> i am developing a use case atm

karl suggests we should be concise and to the point

<cperey> can someone put the URI to the templates into IRC

<tinkster> Karl's template - http://lists.w3.org/Archives/Public/public-xg-socialweb/2009May/0103.html

goal of the templates are to speed up the writing

<cperey> thanks Toby!

templates will give us a common look and feel

Karl is open to comments/modifications re: the templates

hhalpin happy with the user stories templates

<petef> how come petef.a not petef?

hhalpin would like a template for the final report to be put up on the wiki

hhalpin asks if anyone has any experience in this ?

<hhalpin> Does anyone have a final report template as well?

<hhalpin> Does anyone want to take that action - i.e. finding a template for final deliverables?

<tinkster> microformats.org write all specs on mediawiki - perhaps useful?

karl stated that someone should take an action to port the final report template to the wiki

can we not look at other XGs?

<AlexPassant> can a chair close this action -> http://www.w3.org/2005/Incubator/socialweb/track/actions/14

<adam> i can take a stab at it

<hhalpin> ACTION: adam to find a good final report template and port it to the wiki [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action04]

<trackbot> Created ACTION-16 - Find a good final report template and port it to the wiki [on Adam Boyet - due 2009-06-03].

hhalpin: adam seemed to suggest that he would look into the template

Task forces

hhalpin asks if anyone has any comments re: the organisation so we can move onto the task force issue

Task force issues: should we merge context / privacy ?

<hhalpin> Context and Privacy Task Force (Karl Dubost)? For Portability and Architectures Task Force (@@)?

so, harry is asking if people like proposed task force titles

<petef> Portability and Architectures - jsalvachua and petef volunteered last telecon.

cperey: has no preference for times, but thinks we need critical mass

and we need a clear agenda

<petef> signing up to task forces where?

<hhalpin> Maybe we could remind people to sign up for task forces

cperey participation has gone down since the start of XG, cperey thinks we should find out how many people are interested in each Task force

<jsalvachua> petef : we both may start together to push the task force

<karl> karl *MAY* be the task force leader ;)

should we find out how many are interested in each task force

<tinkster> Also interested in portability/arch but not leading.

<cperey> I agree

<hhalpin> Wiki page for each of these task forces?

<petef> where?

<hhalpin> I don't think we do.

<tinkster> I can set up template wiki pages for them.

<petef> I will draft one for portability and architectures

<hhalpin> ACTION: tinkster to draft wiki pages for task forces [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action05]

<trackbot> Created ACTION-17 - Draft wiki pages for task forces [on Toby Inkster - due 2009-06-03].

toby to set up wiki page re task forces, and people should add what they are thinking

portability task force, so hhalpin is interested in how the W3C can promote how data can be made portable

<rreck> i would be happy to contribute but would like to have the discussion on the list

<tinkster> I can just hear typing.

<tinkster> http://www.w3.org/2005/Incubator/socialweb/wiki/Portability_Architecture_TF

jsalvachua: said that he will try and populate the wiki, regarding a roadmap for the portability task froce

<tinkster> http://www.w3.org/2005/Incubator/socialweb/wiki/Context_Privacy_TF

Invited Guests

<hhalpin> 4. Invited Guests

I will put some stuff on the wiki for privacy task force

<hajons> -q

<hhalpin> Anyone else?

invited speaker lists have been populated for the widget topic and the Vcard topic, does anyone else have any ideas on this ?

hajons, asks if people should just write to the wiki, if they want to join a task force ?

<petef> Yes, just sign up to task force on wiki page

or is there some other protocol

<hajons> how do we join the task forces

<hajons> ok

<adam> maybe once the task force pages are created, each person can add themselves as a member

<adam> of the task force

<hhalpin> +1 adam

hhalpin states that we should just add names to the list of task forces on the wiki

<rreck> i will join the privacy task force

http://www.w3.org/2005/Incubator/socialweb/wiki/Context_Privacy_TF

;)

<rreck> i dont feel capable of leading it

are we going to have invited guest for privacy and context ?

harry wodners if there is any mobile interest

from the group

<hajons> yes, I will propose a guest on context / mobile

<hhalpin> tim gave his regrets and daniel applequist can't make this meeting in particular :(

<hhalpin> Perhaps you can explain what the OSLO group is?

<hhalpin> Does it have a web-page?

<hhalpin> Open Sharing of LOcations

christine, is interested the mobile technologies, and christine also contacted someone (?) external, and they are not interested in developing protocols

OSLO group announced start earlier this year, and they are NOT interested with speaking to W3C

<tinkster> Open Sharing of Location-based Objects (OSLO) - http://uk.techcrunch.com/2009/02/27/oslo-accord-pushes-location-sharing-between-social-networks/

<hhalpin> Christine, perhaps stay in touch with them?

<caribou> http://www.w3.org/2005/Incubator/socialweb/wiki/InvitedGuests

there are others which could speak to the w3c

but we need some specific questions

<hhalpin> Maybe brainstorm on the wiki what specific questions would be relevant to these mobile operators?

so that christine could approach people

<hhalpin> Or what is part of their problems?

<hhalpin> What could they need?

nokia people did context and mobile stuff, and I could ask Mor Naaman to talk about the Zonetag project

but i am out of touch with the mobile stuff

harry asked if we could have a wiki page / content regarding the mobile space

<hhalpin> ACTION: cperey to add mobile companies to to Invited Guests and to brainstorm what exact questions or topics would be most interesting [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action06]

<trackbot> Created ACTION-18 - Add mobile companies to to Invited Guests and to brainstorm what exact questions or topics would be most interesting [on Christine Perey - due 2009-06-03].

christine would like an agenda to take to the mobile experts before contacting them

Creating User Stories on the Wiki

<hhalpin> any more use-cases?

there is some work i emailed round from a chap from cambridge

which had some good examples of how privacy in the social web tends to look like

karl thinks that we should add some more user stories, so that we can get a feel for what the XG should be looking at

<rreck> yes i will add a story

<hhalpin> I feel we are still missing some use-cases regarding businesses and developers

<rreck> i wanted to get privacy classes but no one answered my email

ah excellent point

<karl> karl: we should take the current user stories and check them against the actual social networks such as frienfeed, facebook, etc.

<karl> then we can see if our cases make sense

<karl> and then we can identify if there are missing ones.

i posted about this

<rreck> great idea

<tinkster> I've got a developer story - I'll create an action for myself.

harry says that we should have a matrix

<rreck> social network matrix

<tinkster> ACTION tinkster to document developer stories on wiki.

<trackbot> Created ACTION-19 - Document developer stories on wiki. [on Toby Inkster - due 2009-06-03].

showing how social networks uphold privacy

<caribou> almost all the user stories that we have are related to privacy/data protection

<karl> ACTION: karl to create the matix to be filled [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action07]

<trackbot> Created ACTION-20 - Create the matix to be filled [on Karl Dubost - due 2009-06-03].

we have developed at garlik

<rreck> can someone help me find classes of privacy

but it is for government institutions

about what they do with your data

<hhalpin> mischa - could we expland the matrix to deal with commercial social networks?

i could put to a similar oen for social networking sites

<hhalpin> Maybe we could look through alexa to get out the top social networking sites.

<hhalpin> I can do that...

karl says that he will put up a matrix based on the current user stories

<hhalpin> ACTION: To retrieve top X social networking sites from the top 500 sites of Alexa [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action08]

<trackbot> Sorry, couldn't find user - To

<hhalpin> ACTION: hhalpin to retrieve top X social networking sites from the top 500 sites of Alexa [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action09]

<trackbot> Created ACTION-21 - Retrieve top X social networking sites from the top 500 sites of Alexa [on Harry Halpin - due 2009-06-03].

so that people can look at each individual social networking sites

<cperey> Karl, what criteria do you want for this list?

<cperey> the "top" social networks?

<karl> cperey, Alexa traffic

<cperey> irrelevant

<cperey> for mobile

http://en.wikipedia.org/wiki/List_of_social_networking_websites

<karl> cperey, what would be your criteria? :)

<hhalpin> how could we get the top X social networking sites for mobile?

<cperey> Number of unique users per month

<hhalpin> is that list available anywhere?

<cperey> more relevant for all types of social networks

http://en.wikipedia.org/wiki/List_of_social_networking_websites

<cperey> I have a list (not published)

<rreck> wouldnt you just select out of the top 500 alexa sites?

<hhalpin> I thin that list on wikipedia uses the company's own data, right?

there has been some work from cambridge

were they looked into

<karl> I will do Mixi (social network in Japan) because I have an account and they don't open account to people outside Japan

social networking sites

T&Cs

http://news.bbc.co.uk/1/hi/uk/8060407.stm

i would like to invite the phd student which did the work

<cperey> this is a metrics question

<rreck> humming is back

looking for the persons link

<hhalpin> I'm happy to go through alexa if someone else will merge it with wikipedia and Christine's list

<cperey> how do you measure a social network

<hhalpin> this is returning to the metric question

<cperey> OK

<karl> +1 for merging

+1 merging

<rreck> +1 merging

<cperey> 30-50 social networks

<rreck> yes that sounds reasonable

<hhalpin> So we merge list from alexa\wikipedia with Christine's list?

<cperey> by country? worldwide?

<rreck> worldwide

<hhalpin> I would assume world-wide at first, and then later we can break it down by country

hhalpin: asked if we could merge christine's list with the alexa rankign and the wikipedia list so as to pick 30 social networking sites

<hhalpin> unless your data is already broken down by country Christine

<cperey> I can work with Harry on this

<cperey> zakim unmute me

http://www.cl.cam.ac.uk/~jra40/publications/

<melvster> I would suggest you also need IM based networks such as Skype, GTalk, XMPP, if they are not included already, as they have significant usage and maturity

cperey's data is broken down by country

cperey: on/deck and business models.

<hhalpin> 30 or 50?

<hhalpin> Start with 30 and then build if needed?

cperey would be happy with 30

+1

<adam> +1

cperey: will make a list of the last 30

<rreck> top based on number of users?

i would like us to check their Terms and Conditions

and see if they actually abide by it

<hhalpin> ACTION: cperey to make list of top 30 to do profiles on, to merge with hhalpin's list on alexa [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action10]

<trackbot> Created ACTION-22 - Make list of top 30 to do profiles on, to merge with hhalpin's list on alexa [on Christine Perey - due 2009-06-03].

<karl> I'll have to find someone for http://en.wikipedia.org/wiki/2channel

<hajons> Christine, do you list mobile access as a feature too?

<hhalpin> we do it as more of long-term action

cperey: data has information regarding social networking sites, and their features

not terms and conditions

<hhalpin> does your list have the feature?

<hhalpin> feature-criteria like instant-messaging, and all of that?

all of the social networking sites on christine's list are mobile centric

<hhalpin> cperey: "PC-centric" vs. "mobile-centric" and then a continuum inbetween.

<karl> what is a mobile access? specific software for mobile devices?

<rreck> my cell does the same things as my PC

<hhalpin> sounds good to me

cperey: will get her mobile-centric list and we should look at it

<hhalpin> yes rreck, but some sites don't well with mobile phones if you don't have a gphone/iphone/other dataphone

and add an pc-centric social networkings site to the cperey's list

<hhalpin> can we check to see if any of these sites take advantage of user-context, like geolocation from mobile phone?

<hhalpin> Sounds great!

matrix of social networking site

and their features

<cperey> where does this live?

<hhalpin> we should a manufacture a wiki page for the list

we need a wiki page for this list

<cperey> yes, please

<cperey> yes, I will fill in

<cperey> I need to sign off and go to another meeting, bye all

<petef> bye cperey

vcard and portability issues will have its own discussion after this call

<hhalpin> any more comments before we move to vcard?

<hhalpin> such as things we are missing?

does anyone have any issues, there are lots of quiet people about?

have we looked over anything

<tinkster> Dammit. My phone's gone dead.

<tinkster> Still on IRC though.

are we missing anything obvious ?

<rreck> im not sure

this should be an ongoing process

if you think we are missing something

Invited Guest Telecon: VCard in RDF

<hhalpin> 6. Invited Guest Telecon: VCard in RDF

let the mailing list know

<rreck> im trying to find classes of privacy and i cant be the first person to want them

rreck: look at this guys work http://www.cl.cam.ac.uk/~jra40/publications/

<hhalpin> One advantage of the vCard in RDF format (RDF Schema here) is it's use of RDF containers to allow complex vCard expressions. Should this be kept, and if so, how? Which ones, rdf:bag, rdf:Alt, rdf:Seq? How about its use of sub-structure?

I have to leave this call, it is 3

:(

<hhalpin> http://www.w3.org/TR/vcard-rdf

<rreck> mischat: ty

<hhalpin> That's the older vCard in RDF format

<tinkster> Aren't rdf:Bag/rdf:Alt/rdf:Seq generally seen as poor cousins of rdf:List these days?

<petef> I have to leave too, bye

<hhalpin> 1) No use of containers in vCard at all

<hhalpin> 2) Using *only* rdf:List

<rreck> why isnt rdf:bag enough

<tinkster> If the order of items is important, use rdf:List, otherwise don't use a container at all.

<hhalpin> 3) Letting it all be a free for all.

i have to leave this chat now :(

i am sorry

<hhalpin> Someone else can scribe?

right i am sorry i am off now, you are less a scribe now

bye all

<hhalpin> is there use for RDF containers?

<hhalpin> PeterMika: vCard is 99 percent hCard

<Norm> Really? Surely there are gobs of vcards out there never rendered in HTML at all

<hhalpin> PeterMika: vCard in RDF is fairly minimal

<hhalpin> PeterMika: Lots of hCard - between 1-2 billion URLs

<hhalpin> Can you share the stats on the usage of the attribute?

<tinkster> hhalpin: But 98% of those 1-2 billion URLs are presumably on a handful of domain names. Just a few script tweaks could change everything.

<hhalpin> should we merge or not do data-typring?

<hhalpin> so for complex structure is the older version better?

<hhalpin> do we need or/want to substructure?

<hhalpin> I thin Renato had some concern for the subset that wasn't hCard.

<jsalvachua> sorry i have to leave now, sorry, see you.

<hhalpin> And there was a big argument over round-tripping in SWIG a while back...

<hhalpin> timbl: I use this for modelling my addresses

<hhalpin> timbl: these are proper vCards

<hhalpin> timbl: but I can see working with well-defined subset, but would like round-tripping in this subset

<tinkster> There are some parts of vCard which are pretty useless.

<hhalpin> timbl: made contact ontology

<hhalpin> norm: we should stick faithfully to vCard spec

<hhalpin> Here is some of the structuring I think:

<hhalpin> <vCard:EMAIL rdf:parseType="Resource"> <rdf:value> corky@qqqfoo.com </rdf:value> <rdf:type rdf:resource="http://www.w3.org/2001/vcard-rdf/3.0#internet"/>

<Norm> More precisely: I said that if we claim to model vCard, we should model it. If not, we shouldn't claim to be modeling it.

<hhalpin> You CAN do that with the newer hCard

<tinkster> The CLASS property is useless.

<hhalpin> It's just it's a bit confusing because we then use v:EMAIL as a subject, not a predicate

<hhalpin> but we can type predicates

<hhalpin> this I think leads to problems with OWL-DL.

<tinkster> MAILER is pretty useless too.

<hhalpin> But I'm OK with that.

<tinkster> (And has been removed in latest drafts.)

<timbl> CLASS was for what groups it is in?

<tinkster> CLASS has three allowed values: PRIVATE, CONFIDENTIAL and PUBLIC.

<timbl> You don't need a bifg process to make a note obsolete, i think -- jsut change the Status Of This Document.

<hhalpin> no process issue?

<timbl> no process issue.

<hhalpin> So, then we can just merge it with Renato's?

<hhalpin> So, no process

<hhalpin> I would like to NOT have more than one URI for vCard

<Norm> The two are these:

<Norm> http://www.w3.org/TR/vcard-rdf

<Norm> http://www.w3.org/2006/vcard/ns

<hhalpin> Then I would like to put the SIMPLE stuff up front in Renato's, and put more difficult things involving data-structuring and rdf:List towards the end of the spec

<hhalpin> There's also a silly difference in capitalization

<hhalpin> I prefer lower-case

<tinkster> Newer URI comes up #1 on Google for me, searching "vcard rdf".

<tinkster> (without quotes)

<tinkster> Norm's spec==namespace URI.

<hhalpin> I prefer having spec URI == namespace URI and then use conneg

<tinkster> Rennato's namespace = http://www.w3.org/2001/vcard-rdf/3.0#

<hhalpin> I mean, one option

<hhalpin> I mean one option is that we re-use Renato's URI, then use as the namespace URI http://www.w3.org/2006/vcard/ns.

<hhalpin> And if one requests "text/html" for http://www.w3.org/2006/vcard/ns

<tinkster> +1 to timbl's suggestion of marking old one as obsolete and recommending new.

<hhalpin> That's the issue with TR.

<tinkster> hCard GRDDL profile <http://www.w3.org/2006/03/hcard> uses 2006 namespace.

<hhalpin> Well, Norm, I think this is at least part of the community.

<Norm> Fair enough

<hhalpin> What is way forward here?

<tinkster> +1 to just an "obsolete" note, as long as it's clear.

<timbl> Proposed ACTION: Harry to check with Renato he is OK with:

<ivan> action on harry: http://www.w3.org/TR/vcard-rdf would refer to the new version, there will be a 'previous version' link to the current one

<trackbot> Sorry, couldn't find user - on

<tinkster> ACTION hhalpin to http://www.w3.org/TR/vcard-rdf would refer to the new version, there will be a 'previous version' link to the current one

<trackbot> Created ACTION-23 - Http://www.w3.org/TR/vcard-rdf would refer to the new version, there will be a 'previous version' link to the current one [on Harry Halpin - due 2009-06-03].

<hhalpin> that works for me and I think Renato will agree with it.

<timbl> ... http://www.w3.org/TR/vcard-rdf would be kept as the "latest-version" URI, and Reanto's veion woul dbe linke dfrom the new one as a "previous version".

<hhalpin> I guess the other question is we keep the "2006" namespace?

<hhalpin> http://www.w3.org/2001/vcard-rdf/3.0#

<tinkster> My vote: keep both namespaces but only recommend 2006.

<timbl> That is Renato's namespace URI.

<hhalpin> Ivan: suggests namespace URIs that tend to use version causes version

<hhalpin> Ivan: So let's use "2006"

<timbl> I agree that it is unwise to use a version number in the URI

<timbl> The year has no semantics

<tinkster> It's based on vCard 3.0.

<hhalpin> I am also a bit against years in URIs, but that's a minority opinion.

<timbl> But you can use /ns/vcard if you want

<hhalpin> Ivan: would prefer to vCard

<hhalpin> PeterMika: I would agree with "2006"

<timbl> or /ns/pim/adr

<hhalpin> For the time being let's use "2006"

<tinkster> Doesn't that just create yet another URI to include in SPARQL queries, etc?

<tinkster> We already have one too many.

<timbl> Ok, so keep the same 2006 ns

<ivan> http://www.w3.org/2006/vcard/ns

<hhalpin> RESOLVED: keep http://www.w3.org/2006/vcard/ns

<hhalpin> The vCard ontology needs examples. At least one that explains you can attach the properties to URIs for people, not just cards!

<hhalpin> we need a list of examples

<hhalpin> from experience.

<hhalpin> PeterMika: Transforming hCard into RDF

<hhalpin> PeterMika: vCard represents both a person and organization

<hhalpin> PeterMika: hCard is value of organization and fn are the same, the hCard is actually representing an organization and not a person

<timbl> That is not RDF

<hhalpin> PeterMika: The equivalent properties determine type of object

<timbl> So the hcrad > vcard mapping has to do some mapping

<hhalpin> PeterMika: so address and whatnot can all apply to person

<hhalpin> PeterMika: AND organization

<hhalpin> PeterMika: Does person have vCard etc.

<hhalpin> PeterMika: And then the vCard have an address etc.

<hhalpin> PeterMika: These are two main points people struggle with

<hhalpin> TimBL: The last one is the major one.

<hhalpin> TimBL: Are we modeling a file or person

<hhalpin> PeterMika: The documentation leads this question open

<hhalpin> I'm noting unclarity about this is WHY there's no examples :)

<hhalpin> I could not consensus on this.

<timbl> I agree that one should moddl the person not the card.

<timbl> Like you model a book, not a library card.

<adam> yes

<hhalpin> toby: 4.0 includes a property "kind" that demarcates between people organization and group

<hhalpin> +1 vCard 4.0

<hhalpin> timbl: which would be a functional mapping to a RDF class

<hhalpin> toby: individual or pre-defined group or organization

<hhalpin> toby: also maybe one for "place"

<timbl> Those shoudl defintely map to classses.

<hhalpin> how stable is vCard 4.0?

<hhalpin> Should we track it?

<tinkster> Not especially stable.

<tinkster> yes, certainly - should be able to attach these to Person/Organisation URIs.

<hhalpin> ivan: hhalpin said having the same person with several vCard is an edge-case

<hhalpin> ivan: since I have two addresses

<hhalpin> ivan: so in my phone I have two entries for my name

<tinkster> I think vCard 4.0 drafts have ways of representing multiple sets of contact information in one card.

<hhalpin> I am liking vCard 4.0 :)

<tinkster> e.g. this phone, this fax and this address are for one set of uses; and this phone and this address are for another.

<tinkster> I've not really studied that part of the syntax though.

<hhalpin> timbl: so you'll miss the fact that these two vCards are not the same in RDF.

<uldis> a person may have different vCards which they "give" to people same as one may have different versions of a business card

<hhalpin> timbl: about the same person

<hhalpin> timbl: does this mean ontology is broken/

<hhalpin> ivan: vCard represents me or my address, I think it represents my address

<hhalpin> timbl: we do not walk about what a vCard represents

<tinkster> [ a vcard:Vcard] vcard:sameOwnerAs [a vcard:Vcard ] .

<tinkster> (No, vCard doesn't have a sameOwnerAs property, but we could always define such a term.)

<hhalpin> so we don't drop the vCard class

<hhalpin> we keep it and keep domains pretty open-ended

<libby> I think that's because the spec's a bit ambivalent norm

<hhalpin> but in our examples we use People and Organizations

<hhalpin> This would make it clear to users, since most users will just look at examples

<hhalpin> timbl: we should be default not give it any class.

<hhalpin> should we add this to the GRDDL and the spec, this weird hCard algorithm?

<hhalpin> to determine people and organization?

<pmika> if fn=org => organization is not possible to express in OWL

<hhalpin> so it should be in GRDDL?

<hhalpin> Not express it in OWL, but *mention* it in RDF spec and then implement it in GRDDL.

<pmika> yes, it has to be in the hcard-to-rdf conversion

<hhalpin> we should make a test case here

<hhalpin> org class only has name and unit.

<hhalpin> PeterMika: So these properties should be extended to organization class

<hhalpin> PeterMika: ALL properties can be extended to organization class

<hhalpin> PeterMika: for example, "adr". Strictly, in vCard people have addresses,not organizations, but people use addresses directly on organizations in hCard.

<tinkster> contact:SocialEntity ~= foaf:Agent.

<tinkster> orgname, orgunit

<tinkster> orgunit can be repeated.

<tinkster> [ a v:vCard ; v:fn "Tim Berners-Lee" ; v:org [ a v:Organization ; v:organization-name "MIT" ; v:organization-unit "CISAL" ] ]

<tinkster> is how it currently works.

<hhalpin> OK, am a bit confused.

<tinkster> Organisations are messy in vCard RDF because they're messy in vCard.

<timbl> org MIT unit CSAIL means memberOf [ a Unit; name "CSAIL"; partOf [ a Org; name "MIT"]]

<tinkster> There are organisation properties which "hang off" people, plus the convention of fn==org whih means that the entire vCard represents an organisation.

<hhalpin> PeterMika: allow both, have examples for both cases

<timbl> org MIT means memberOf [ a Org; name "MIT"]

<hhalpin> PeterMika: show an organization where the unit is not used, just give it name and some other properties.

<hhalpin> PeterMika: A case where the vCard is using to describe an organization

<tinkster> FOAF's model for people, orgs, membership is a lot more sensible.

<hhalpin> TimBL: You need to spot these patterns

<hhalpin> PeterMika: Yes, you need to do that in transform.

<hhalpin> Perhaps we can add that to GRDDL.

<tinkster> foaf:member

<hhalpin> Norm?

<timbl> Has FOAF got org is part of biggerOrg?

<hhalpin> skos:widerThan :)

<tinkster> no, but dublin core has "hasPart/isPartOf"

<libby> foaf:Organization

<timbl> I wonder whether a gorup is a SocialEntity

<libby> don;t see any properties tho

<AlexPassant> http://xmlns.com/foaf/spec/#term_Organization

<libby> "This is a more 'solid' class than foaf:Group, which allows for more ad-hoc collections of individuals. These terms, like the corresponding natural language concepts, have some overlap, but different emphasis. "

<hhalpin> ok

<libby> not sure if there's any formal subclassing

<libby> probaby not

<timbl> Maybe socialEntity should be explicitly allowed to incldeu a group.

<libby> of group, that is

<AlexPassant> may be used together with foaf:member ? (":csail foaf:member :mit")

<libby> not very clear

<tinkster> AlexPassant: not sure of :csail foaf:member :mit.

<timbl> Good Q Alex

<uldis> re. earlier mention of SIOC - for representing organisations FOAF would be more appropriate that SIOC

<pmika> there is no equivalent of organization-unit in FOAF and I would not be in favor of bringing in a single FOAF class into the VCard spec

<tinkster> :microsoft foaf:member :w3c .

<timbl> FAOF more appropraite thatn SIOC?

<hhalpin> Re FOAF and VCard, I think we first fix vCard RDF spec, because that's relatively easy, and then see what the future of FOAF is in the next telecon.

<libby> sounds good

<AlexPassant> previous thread on foaf:Group / foaf:Organisation http://lists.foaf-project.org/pipermail/foaf-dev/2007-January/008396.html

<Norm> Good luck to all! See you on the next telcon

<ivan> thanks to norm

<libby> cheers AlexPassant

<libby> did you get a response? doesn't look like it AlexPassant

<libby> perhaps bump the thread?

<AlexPassant> libby: unfortunately, no answer - I sent a similar one a year later but no answer as well

<hhalpin> republish it for next week?

<hhalpin> Formal note, it's an IG note, not a SWXG note.

<hhalpin> XGs can't do Notes even :)e

<hhalpin> send us examples

<hhalpin> PeterMika will examples.

<hhalpin> Meeting adjourned

Summary of Action Items

[NEW] ACTION: adam to find a good final report template and port it to the wiki [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action04]
[NEW] ACTION: AlexPassant to see about SIOC liason [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action01]
[NEW] ACTION: cperey to add mobile companies to to Invited Guests and to brainstorm what exact questions or topics would be most interesting [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action06]
[NEW] ACTION: cperey to make list of top 30 to do profiles on, to merge with hhalpin's list on alexa [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action10]
[NEW] ACTION: hhalpin to retrieve top X social networking sites from the top 500 sites of Alexa [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action09]
[NEW] ACTION: karl to create the matix to be filled [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action07]
[NEW] ACTION: tinkster to draft wiki pages for task forces [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action05]
[NEW] ACTION: To retrieve top X social networking sites from the top 500 sites of Alexa [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action08]
 
[CONTINUES] ACTION:danbri to find someone from Opera to talk Widgets
 
[DONE] ACTION: danbri sketch a 5 line template for interaction with other groups (cf InvitedExperts, DiscussionTopics) [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action02]
[DONE] ACTION: karl to produce a template for TF deliverables. [recorded in http://www.w3.org/2009/05/27-swxg-minutes.html#action03]
[DONE] ACTION: AlexPassant to create a liasons list on the Wiki and list himself for SPARQL.
[DONE] ACTION: hhalpin and danbri to invite guest speakers and investigate their schedule.
[DONE] ACTION:danbri to document throwaway IDs on Wiki
[DONE] ACTION:oshani to document takedowns as a user story
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2009/05/27 16:34:14 $