Minutes: XDI TC Telecon Thursday 1-2PM PT 2010-06-10

[Drummond Reed <drummond.reed@xdi.org>]

Following are the minutes of the unofficial telecon of the XDI TC at:  

Date:  Thursday, 10 June 2010 USA
Time:  1:00PM - 2:00PM Pacific Time (21:00-22:00 UTC)

ATTENDING

Giovanni Bartolomeo 
Joe Johnston 
Joseph Boyle
Drummond Reed 
Markus Sabadello
Bill Barnhill 


AGENDA

1) NEWS

 * Drummond conveyed that he had a meeting with a large data aggregator in which he did a deep-dive on XDI to show them how it could be applied to personal data sharing using the PDX vocabulary. They were very appreciative of the power of XDI and why the graph model differs from RDF.
 * Drummond also reported that the Higgins Project, where the current XDI4J Project is hosted, has added a weekly 30-minute "Mapping Higgins Context Data Model (CDM) to XDI". That call will be weekly on Thursdays from 10-10:30AM PT. It will be on the same bridge as the XDI TC call.
 * Bill said he plans to publish this weekend the JSON Schema of the JSON format he is using in his code.
 * Giovanni reported that an journal article about the integration of XDI and XDM (XML data management) that was accepted a few months ago (about which he sent an email to the list) has now been issued. It is not on the Web, but if you are interested in a copy for personal use, let Giovanni know. http://www3.interscience.wiley.com/journal/123206047/abstract?CRETRY=1&SRETRY=0

2) PDX EXAMPLE DOCUMENT

   http://wiki.oasis-open.org/xdi/PdxExample

The $is topic is currently being discussed on the list. The only other topic we had time for on today's call was the relationship between PDX and FOAF

Giovanni asked how we will proceed with this type of mapping, i.e., creating an XDI dictionary where the majority of the job is mapping to existing RDF vocabulary. Drummond gave the following example of what such a vocabulary mapping would look like:

    $(http://some.example.com/rdf-vocabulary-uri)
    
Giovanni wondered why the leading global context symbol was necessary. Bill and Drummond replied that it makes the cross-reference into an absolute XRI.

http://xmlns.com/foaf/spec/#term_familyName <-- FOAF family name definition

(http://xmlns.com/foaf/spec/#term_familyName) <-- relative XRI

+(http://xmlns.com/foaf/spec/#term_familyName) <-- absolute XRI

 Bill's POV was that this is needed because 

(http://xmlns.com/foaf/spec/#term_familyName)

cannot be differentiated in type from

(+foo)

However 

+(http://xmlns.com/foaf/spec/#term_familyName)

is differentiatable from

(+foo)

and we can say any XRI of the form +(xxx) maps to the URI space

Giovanni:
Not all URI associated to ontology concepts are "+" (i.e. things), they could be organizations "@", individuals "=", and "meta concepts" "$" as well.

Bill:
+(http://example.org) is not the XRI for example.org, that'd be @example. In other words, +(http://example.org) points to *the resource* identified by http://example.org

Giovanni:
What about...

@example/$is/(http://example.org)

Are they not  just synonyms?

Bill:
Wouldn't it be 

@example/$owns/(http://example.org)

Not $is?

Giovanni:
Interesting, there is a discussion in the RDF/OWL community on how to differentiate URI pointing to resources from concepts.. it is the discussion about cool uri, etc... maybe you know as well. I've to check again their pages and come back to this topic then."

Bill:
Yes, that's what I am drawing from. If I have a URI for http://drummond.reed.name does that identify Drummond or does it identify a resource containing data about drummond. "Cool URIs Don't Change" was the paper that partially sparked it I think, TBLs paper.
TBL = Tim Berners-Lee

Giovanni
right

# ACTION: take a look at that paper and check their solution, then compare with ours

Drummond:

I strongly recommend the following paper for the W3C RDF 2.0 Workshop (http://www.w3.org/2009/12/rdf-ws/) later this month:

“When  owl:sameAs isn’t the Same: An Analysis of Identity Links on the Semantic  Web”, by Harry Halpin, Ivan Herman, and Patrick J. Hayes

It has a great deal to say about "identity links" and representations.

----

We did not reach the following items on the PDX Example Doc question list:

- associativity
- "complex predicate"
- usage of "relative" ref e.g. (/+postal+address)
- "how much" goes in the subject and "how much" in the predicate...
- $ as a suffix



3) OTHER EXAMPLE PDX SCENARIOS

We did not reach the  other example PDX scenarios Joe Johnston would like to document. His  list included:

-  Social address book
-  Photo sharing
-  Friending
-  Chatting
-  Event sharing
- Calendaring
-  Wiki

# DRUMMOND to create an XDI TC wiki page for Joe to populate with his list and others to start documenting the mappings.


3) LINK CONTRACTS

We did not have time to tackle the following - we will continue on the next call:

Continue discussion from the list on the following special questions  related to link contracts (but not the link contract format itself)  raised by Markus:

-  If link contracts are enabled at an XDI endpoint and no link contracts  exists, what is the "default" behavior? Allow nothing? Allow only $get?
- Can an XDI endpoint have a special notion of the XDI endpoint  "belonging" to a user?
- Is yes, does that user automatically have  "full rights" to the XDI endpoint outside of normal link contract  evaluation?
- As an alternative to signing XDI messages with a key, can an XDI  message contain a user's password for authentication?


5) NEXT CALL

The next call is next week at the regular time.