Minutes: XDI TC Telecon Thursday 1-2PM PT 2009-01-22

["Drummond Reed" <drummond.reed@cordance.net>]

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

 

Date:  Thursday, 22 January 2009 USA

Time:  1:00PM - 2:00PM Pacific Time (21:00-22:00 UTC)

 

ATTENDING

 

Tatsuki Sakushima

Markus Sabadello

Mike Mell

John Bradley

Drummond Reed

Giovanni Bartolomeo

 

REGRETS

 

Nick Nicholas

 

 

AGENDA

 

1) CONFERENCE CHAT ROOM

 

We started using a chat room to share pointers and keep a question queue:

 

      http://webconf.soaphub.org/conf/room/xri

 

 

 

2) RATIONALE OF XDI METAGRAPH MODEL

 

RE Drummond's message posted here...

 

      http://lists.oasis-open.org/archives/xdi/200901/msg00058.html

 

...we briefly discussed whether there were: a) any questions about the rationale behind the metagraph model, or b) any significant issues remaining with it besides completion of the RDF mapping of each metagraph predicate. Among those on the call the answer to both was no, however we must loop back with the rest of the TC members.

 

On a related topic, John made the point that the metagraph statements only describe parts of the graph, so if you want to do semantic reasoning about the graph, you don't want to do it on the metagraph. Drummond suggested that for the XDI 1.0 specs we could simply specify a two-step process for doing semantic reasoning with any XDI RDF document:

 

      Step 1: Transform all metagraph statements into regular graph statements (i.e., remove all metagraph predicates).

 

      Step 2: Transform the remaining XDI RDF graph into conventional RDF.

 

This way standard RDF reasoning tools can operate on the graph without modification into XDI RDF tools.

 

 

3) GLOBAL GRAPH MODEL PROPOSAL

 

Giovanni posted a proposal dealing with syntactic and semantic

correctness of XDI documents in a global graph model:

 

      http://lists.oasis-open.org/archives/xdi/200901/msg00060.html

 

We spent the rest of the call reviewing this with him to understand the key points of his proposal.

 

Giovanni's first point was that metagraph statements may limit or impose specific constraints on the regular statements you can have in the graph. That's why a graph must be checked for syntactic and semantic correctness.

 

In Giovanni's notational model, all native XDI predicates (including the metagraph predicates) imply that the subject exists in that context. So his model shows these relationships explicitly (with all the metagraph nodes in red). Drummond suggested that just using red arcs (or in fact different colored arcs for each metagraph predicate) might be simpler.

 

There was conversation about whether all inverses need to be explicitly stated in the graph. Drummond suggested that the inverses of the metagraph statements do not need to be explicit as they derive directly from the metagraph. However all other inverses (regular inverses) must be explicit.

 

The final part of our discussion was about contexts. Giovanni's graph notation shows contexts explicitly, with the arcs to each subject representing the XRI of the subject. This notation is also able to show subcontexts. There was some confusion about $/$is statements in relation to subcontexts stemming from the V11 XDI RDF Model doc (http://wiki.oasis-open.org/xdi/XdiRdfModel). Drummond took the following AI:

 

# DRUMMOND to post a V12 bringing it current to the metagraph model posted on the wiki at http://wiki.oasis-open.org/xdi/XdiOne/RdfGraphModel.

 

We agreed to continue the conversation on the list.