See also: IRC log
<trackbot> Date: 22 September 2010
I'm just bad at scribing. it's a terrible failing, but the pain of what I produce is often worse than nothing.
I can but try
<melvster> paul's slides: http://www.slideshare.net/ptrevithick/higgins-active-clients-and-personal-data-stores-v2
<hhalpin> PROPOSAL: accept minutes http://www.w3.org/2010/09/15-swxg-minutes.html
<hhalpin> ACCEPT: minutes as true record http://www.w3.org/2010/09/15-swxg-minutes.html
<hhalpin> Meeting: Next meeting Mike Hanson + Chris Messina on browser as social agent, final meeting
hhalpin: wants to broadly review final report...
hhalpin: rough draft is in usual
place; HTML export not done yet
... trying to shorten and lengthen :-)
... adding XRD, pasteform, OAuth, OpenID, other coverage
hhalpin: suggests XAuth, OExchange, Microformats, P3P / privacy rulesets
hhalpin: conclusions wants heavy attention
<paul> Personal Data Spaces aka Personal Data Stores aka Personal Data Lockers aka Personal Data Banks
hhalpin: tried to filter concerns around RDFa into final report "hard to find URIs for things"
(that's an RDFa concern?)
scribe: pay much attention to
proposal for W3C to rejigger incubator group proces...
... has shortened RDF stuff
... informal rule "each topic we've discussed should get 2 paragraphs"
<hhalpin> So you would want more RDF in the report
mischat: massive RDF advocate,
still felt that final report was RDF heavy, concerned that it
*may* scare people away
... identity section was pretty much all about RDF; profile data and SIOC stuff was all RDF...
... should be talking about RDF as the future, but whole document shouldn't be about it
<melvster> im an rdf fan
MacTed: arbitrary limits like that bother me
hhalpin: working toward
concision, would like to receive cut-and-paste replacement
... other controversial decision: choice between Renato's and Henry's graphics
... Renato's are prettier and more contemporary in appearance; Henry's appear to be more technically correct/complete
... will try to wrap things up next 2 days
... group shuts down in a week, so responses are important now
<mischat> hhalpin: we do have a bit on higgins :)
paul: updated deck ---
... all data about a person comes from own head -- self-assertions aren't trustworthy
... people don't exist on the internet -- it's all about browsers and documents
... what does this mean for people?
... pour their information out, and give up all rights to data (which better not be copyrighted) on sites
... (of course, exaggerating but with purpose)
<hhalpin> Moving verified attributes from A to B
paul: can't prove age online,
unlike in liquor store
... (slide 7) missing piece!
... identifiers are just more attributes
... important thing is what are you trying to say
... (slide 8) "cards" hold purpose/site-targeted sets of attributes
... managed in (various terms) wallet/selector/etc.
... (slide 10) putting Higgins in perspective against other similar/parallel projects (e.g., Microsoft)
... (slide 11) user control is key. "digital identity" is very broad -- anything that might be in a FOAF file
... (slide 12) rough timeline shows convergence and focus
... (slide 13) "why cards are good"
... most of shown cards reflect real projects (not AmEx)
... (slide 14) personal = self-assertions. managed = verified, or assertions made by others
... (slide 15) straw-man UI sample
... no per-relying-party login, akin to OpenID, WebID, etc.
... (slide 16) may send subsets of attributes even from a single card
... (slide 17) mix-and-match interoperability
<hhalpin> *lots* of people
paul: (slide 20-22) info &
... (slide 23) good news -- lots of involvement and interest
... work-in-progress -- still looking for lighthouse big (or medium) size relying site(s)
... (slide 24) refactoring project and organization based on experience since inception
... (slide 26) not yet released, but in the works.... Higgins 2.0
... users don't like seeing everything that's happening, they'll just click to close the windows
... bits-and-bytes are good, but legal/trust frameworks are also necessary
... working to further reduce required user interaction
<hhalpin> overloading is a bit confusing
<hhalpin> too many identity standards to even track!
paul: working on "better with"
more than "better than"
... end-user wants solutions that work in the airport kiosk when you don't have an active client, but will accept that it works better when *do* have active client
... adding more card types
... (slide 30) Personal Data Store is not monolithic silo, but integrates all person's data from all their sites...
... (slide 31) Higgins 2.0 jumps into ontology/vocab use, embracing and extending with card metaphor
... card (or set of cards) for Amazon, or for work-related, or for address info, etc. forming persona
... URL for Persona Data Model (can't be cut-and-paste, won't transcribe now)
... (slide 34) potentially adding procedure to data
... (slide 36) dashboard here might be better called wallet...
<hhalpin> Fabien Gandon also uses the term "wallet" in his research
paul: rest of deck is technical detail appendices
hhalpin: question time....
... wants a couple paragraphs for final report
... looking for "meta-identity" view
... building OpenID atop OAuth involves server-side redirect
... has potential for phishing attacks
... Infocard seems to handle that?
paul: yes. phishing is mitigated by some smartening of the browser
hhalpin: Infocard seems a near relative to Henry Story's WebID
<melvster> WebID is a 'hub of identities'
paul: there seems to be some
... world is contextual. we should be able to have identifiers with minimized scope.
<hhalpin> so the big difference "multiple" personae
<melvster> many WebID's are linked via owl:sameAs
paul: presumption is that context
... contrasted with OpenID where personae *can* be done but are not treated as desirable
hhalpin: 1. great if browsers could implement this. seems like Infocard has got pretty good buy-in.
paul: doesn't belong in the
browser. this belongs above the OS, and below the browser, but
tightly integrated throughout
... working on making it seem like a browser extension, even though it's "really" not
... there are things other than browsers, still
<hhalpin> below app, in-between OS
<hhalpin> and browser
paul: belongs below "app" as well as browser, app is whatever acts on the network
hhalpin: what are people going to be interested in after HTML5?
paul: Mozilla's view is
Mozilla-centric, which is OK, but this isn't necessarily good
for the end-user
... multi-browser, multi-app world means we need standards to help the user who wants their "stuff" synched across browsers/apps
hhalpin: there seems to be some
client-side ID space starting to take shape that could work in
harmony with server-side ID space
... there's no making browser vendors do anything, but they can be encouraged
... EU is in middle of formulating plans on privacy and identity requirements
paul: Europe and Canada will be
very interested in active clients, possibly moreso than
... Infocard is 3 things -- metaphor, architecture, & bits-and-bytes
hhalpin: W3C can do various things.... and ID space may be ripe for W3C participation
(I have to go soon)
<hhalpin> so two workshops
<hhalpin> active client...
<pchampin> any SWXG event planned at TPAC?
<pchampin> I'll be there. can't wait :)
<hhalpin> I'll send e-mail
<hhalpin> thanks MacTed!
I know there are things to do generate minutes....
but not what they are
<hhalpin> Wish we had bblish on the phone
<hhalpin> trackbot, generate minutes
<trackbot> Sorry, hhalpin, I don't understand 'trackbot, generate minutes'. Please refer to http://www.w3.org/2005/06/tracker/irc for help
<hhalpin> don't worry MacTed, your job is over!
<hhalpin> trackbot, end meeting
This is scribe.perl Revision: 1.135 of Date: 2009/03/02 03:52:20 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/13/15/ No ScribeNick specified. Guessing ScribeNick: MacTed Inferring Scribes: MacTed Default Present: +1.781.416.aaaa, MacTed, hhalpin, mischat, melvster, +1.781.392.aabb, +1.760.705.aacc, pchampin Present: +1.781.416.aaaa MacTed hhalpin mischat melvster +1.781.392.aabb +1.760.705.aacc pchampin WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth Found Date: 22 Sep 2010 Guessing minutes URL: http://www.w3.org/2010/09/22-swxg-minutes.html People with action items:[End of scribe.perl diagnostic output]