[higgins-dev] Notes from today's call

[higgins-dev] Notes from today's call

Logistics

Attendees

Agenda

Left hand nav: changed "community" to "developer" per eclipse stds (e.g. see [8]); and eliminated "sub" items. http://wiki.eclipse.org/Developer - completely revised

1 remaining: https://bugs.eclipse.org/bugs/show_bug.cgi?id=317227
Valery: we've been having some build problems; also the downloads page.
Mike: do you think we can get this done for July 1
Paul: I'll update the project plan accordingly
Have you reviewed this? [7]
Paul: if you could update this page for next week that would be great.
Paul: Also if you could review http://wiki.eclipse.org/Developer_Team_Page#Building section too

Go to [6] click top link
13 open items (Mike, Alexander, SergeyL) - we're not going to do some of these
Mike: Azigo is going to contribute a new implementation of the STS
Paul: Are we going to do this for 1.1?
Mike: It will likely take 2 weeks to get the work done. I think we're better off leaving this to 2.0?
Paul: Can you review and push to 2.0 a number of these open bugs currently in 1.1M9
Mike: the code for most of the solutions is done; this is mostly about documentation; Most of these involve wiki pages changes and build process changes, etc. Once one of them is done, the rest should fall pretty quickly behind it.
Valery: yes, I agree.
Mike: Irina has been making progress.
Mike: Let's delay the date to 7/22 (one week)

Brainstorming Q&A session with David
I'm working on a project in the health-care space.
It involves integration with salesforce.com.
Could we design it so that the individual patient (and doctors) have their own identity and can query attributes of their own personal data store, they can have have visibility into where the data comes from.
People, esp. WRT medical records, really want a sense that they are in control. One of the main drivers of this is to give the individual patient has an audit trail. What organization accessed what data when and under which policies.
I'm looking to learn more about Higgins.
David: are there other initiatives in the identity industry around giving.
Paul: should look at: http://kantarainitiative.org/confluence/display/healthidassurance/Home

Harmonization: continue to work through [3]
Paul explained that h:correlation is different from the stronger owl:sameAs
Joe: asked if $is and it's use to correlate i-name, i-number, etc. is that the same as h:correlation?
Paul: I think that's closer to h:synonym
Joe: in general XDI tries to use fewer verbs
Paul: Higgins too. We try to invent as few as possible. Not only that but we only use a subset of rdf, rdfs, and owl so as to make it easier for context provider implementers who are NOT using rdf technology.
Joe: the impression I get of XDI is that some distinctions are made by surrounding context
Paul: the implication of analysing surrounding context is that performance (during mapping) would be reduced
We updated [3] as we discussed $is$a vs. rdfs:type and rdfs:subclassOf as well as XDI comments (in non-JSON-serializations) vs. rdf:comment

Next call in two weeks (July 8th)

--Paul