Minutes: XDI TC Telecon Thursday 1-2:00PM PT 2010-11-18

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

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

Date:  Thursday, 11 November 2010 USA
Time:  1:00PM - 2:00PM Pacific Time (21:00-22:30 UTC) - SEE NOTE ABOVE

ATTENDING

Markus Sabadello 
Drummond Reed
Giovanni Bartolomeo 
Joseph Boyle

REGRETS
Joe Johnston 
Michael Schwartz


THE GOTOMEETING FOR TODAY IS:
     https://www2.gotomeeting.com/join/969244355

THE IDEARPAD LINK FOR TODAY IS:
      http://xdi.idearpad.org/13

Please try to preface each of your comments with your name so the transcription into the minutes is easier.


1) DISCUSSION OF TIMING AND OPEN ISSUES FOR FIRST WORKING DRAFT SPECS

See Giovanni's summary of open issues:

   http://lists.oasis-open.org/archives/xdi/201011/msg00019.html
   
First, Giovanni invited all TC members to review the two areas where his email indicates we are close to agreement.
   
We reviewed the spec list at:

  http://wiki.oasis-open.org/xdi/XdiOneSpecs
  
Key points:

 * We need the graph model and serialization to be first and the most mature since the rest of the specs depend on these. The graph model is normative for the graph structure, and the Serialization spec defines the different normative serializations of the graph model.
 * The REST and Async API specs are also critical for defining the operation of XDI as a protocol. Joseph suggested that we may want to factor the API and Synchronization specs differently. 
 * Joseph suggests that they fall into two buckets: dynamic and static. Static: link contracts, dictionaries, maybe signatures. Dynamic: APIs, synchronization, discovery.
 * We discussed the Context Discovery spec, and agreed we need to wait and see whether it will be needed, or whether we will be able to use the XRI 3.0 Resolution spec (which is also an outstanding action item for the XRI TC).

# DRUMMOND: Move Link Contracts up in the priority list. DONE.


2) ISSUES TRACKING/CLOSURE

We discussed the options for issues tracking and closing. The consensus was to use a lightweight wiki-base issues approach for early stage and general issues, and graduate them to JIRA when they are mature and more mechanical in nature.

# DRUMMOND to start XDI 1.0 Issues Tracking page.


3) DISCUSSION OF XDI MODELING OF PERSONAS

See Giovanni's message at:

  http://lists.oasis-open.org/archives/xdi/201011/msg00020.html
  
Drummond sent a response to Giovanni's message that explained the differences in the two patterns he referenced, and gave some examples of supercontexts (roles) and subcontexts (personas).

Giovanni noted that the final examples Drummond gives do not have sufficient semantics for it to be distinguishable that the predicates (e.g., +home, +work, +baseball) are describing personas.

We discussed that the semantics of a predicate whose object is a set of XRIs, possibly constrained by the range of the predicate, e.g., +friend takes a set of XRIs identifying resources that are people (similar to owl:allValuesFrom), is a pattern we already support, but which is not sufficient to capture the semantics we need.

# DRUMMOND, GIOVANNI (and anyone else): prepare proposals for semantics that would be clear and consistent for describing personas.

Giovanni made another point about the examples in Drummond's message. The first set...

@company=alice
@sports.club=alice

...do not express anything explicit about the relationship of the two XRIs, whereas...

@company+salesperson=alice
@sports.club+pitcher=alice

...express the relationship clearly. So Giovanni suggests that the first two should be:

@company+member=alice
@sports.club+member=alice

Drummond points out that the XDI metagraph symbol proposal has a generic way of expressing "membership" anywhere in the graph:

@company()=alice
@sports.club()=alice

Giovanni made the point that our semantics with regard to mereological relationships must be very precise.

In particular, the mereological relationship ("part of") should be a wider concept than the memberhip relationship ("member" in role relationships), as the former models also the relationship between a subject and its attributes. So open question is whether () is the proposed symbol for a built in operator with the semantics of "part of" or only just for "member".

Drummond said that in preparing his persona analysis, he had a fascinating realization about the semantics of the $ symbol. It can be summarized this way: if

+foo$1
+foo$2
+foo$3

all express instances of +foo, then +foo$ expresses the set of all of these instances.

This ends out being the precise metagraph semantic equivalent of the plural form of a noun in the English language. For example, you can request all the instances of =alice's email address (which in English woudl be "Alice's email addresses") as:

=alice+email$

Giovanni asked: what about =alice+email -- what does that identify differently in the graph? Drummond answered that it returns the entire graph rooted on that node, which may be other metadata about Alice's email addresses, not just the precise set of them.

# DRUMMOND to write up some further examples as a page on the wiki.


4) NEXT CALL

Next Thursday is Thanksgiving in the U.S., so we will not hold a call.


------------
ONGOING ISSUES LIST

Each of these is a candidate for the agenda for future calls.

* PROPOSED CONSTRUCTS/OPERATORS FOR XDI

Discuss the following wiki page originally posted by Giovanni:

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

* DICTIONARY STRUCTURE

Mike would like an example of the PDX dictionary as soon as we can do it.

*   EQUIVALENCE SEMANTICS

Close  on whether we need an additional $ word   that is the equivalent of  Higgins Personal Data Model (PDM)  semantics   of h:correlation, which  is not as strong as $is.

      http://lists.oasis-open.org/archives/xdi/201006/msg00036.html

* COOL URIS

Continue previous discussion about the use of standard RDF URIs in XDI:

  http://lists.oasis-open.org/archives/xdi/201006/msg00023.html