Minutes: XDI TC Telecon Thursday 1-2:00PM PT 2010-12-02

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

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

Date:  Thursday, 02 December 2010 USA
Time:  1:00PM - 2:30PM Pacific Time (21:00-22:30 UTC)

ATTENDING

Michael Schwartz
Giovanni Bartolomeo
Joe Johnston
Drummond Reed
Markus Sabadello
Bill Barnhill


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

THE IDEARPAD LINK FOR TODAY IS:
     http://xdi.idearpad.org/14


1) DEMONSTRATION BY MIKE SCHWARTZ OF GLUU LDAP NAMESPACE

Mike demonstrated how he has incorporated XRI i-numbers into the GLUU
LDAP namespace design to make the entries globally addressable.


2) DISCUSSION OF CORE DECISIONS FOR XDI 1.0 SPECS E.G. WHERE DO WE GO
FROM HERE ON SEMANTICS?

Discuss the thread about the tension between the pragmatic need to
publish the 1.0 specs and the purist need to get the semantics right.
The thread ends with this message (read backwards from there):

    http://lists.oasis-open.org/archives/xdi/201011/msg00042.html

In specific, discuss the two points -- the XDI/RDF relationship and
the XDI metagraph model -- in Drummond's message to the list on
Monday:

   http://lists.oasis-open.org/archives/xdi/201011/msg00043.html

Bill started the discussion with asking about, "Where does the TC
stand with regard to semantics?"

Drummond said that his answer is captured in this email. He summarized
it this way:

    1) He believes the metagraph model
(http://wiki.oasis-open.org/xdi/MetagraphSymbols) is a solid
foundation semantically.
    2) He believes that we should press forward with the 1.0 specs
based on this model even if we have not figured out the more complex
or advanced semantics (e.g., negation).

Giovanni said that XDI must guarantee some very basic semantic
features, which are still not clear in the current graph model. He
believes that these need to be clarified before publishing the specs.
He started a wikipage intended to illustrate the basic constructs
(http://wiki.oasis-open.org/xdi/XdiNewFoundation) he believes XDI
should offer, and how these constructs can be combined to form a
minimal but solid formal model for XDI.

# ACTION: Giovanni to identify any outstanding semantic issues and
document them on that page.

Mike said he believes we should proceed with what we have figured out
as long as it can be implemented and tested in order to get to market
as quickly as possible. He feels the power of XDI begins with the
ability to give an object on the Internet a globally addressable
identifier together with a standard way to look up information about
it, and that this capability is something that should be implemented
as soon as we can.

Giovanni agrees with what Mike is saying, but feels that those
capabilities alone are "not enough". These things can be done with
URIs and XML today, for example XCAP. So he feels that if this was the
only features we have for XDI, then the Internet community will reject
it.

So the reason to have some very basic semantic capabilities within XDI
is essential to achieving market success. An example is having a
standard way in which to query an XDI document using XDI built-in
reasoning capabilities over XDI graphs [see also debate about SPARQL
and XQuery]. He states the goal this way: "The goal of XDI should be
able to let implementers do basic semantic operations on an XDI graph
more easily than they can with RDF."

Markus agrees with both points: he wants to be able to push forward
with implementations, but he appreciates the value that semantics
brings, and that it would be a good feature to support in XDI.

Joe admits he doesn't know enough about the semantic side but he is in
favor of moving forward with adoption.

Drummond then asked a second question: "Is everyone in favor of
producing our first Working Draft specifications?"

Bill said that he's in favor of doing that with the caveat that we be
careful to emphasize that it is still at the Working Draft stage, and
should not be referenced or implemented against outside the TC.

Giovanni said that it is okay to move to a Working Draft stage with
the caveat that if there are issues, this Working Draft becomes the
place that we focus discussion on the issues, and not a way to avoid
any issues.

Drummond also brought up his point from his email about the mapping of
XDI to RDF. He asked if we are at the point of having an answer to
this longstanding question.

Bill pointed out that "RDF is only one flavor of expressing
semantics", and so just because you can map to RDF is not proof that
you have a strong semantic model. He pointed out that N3 is actually
more expressive than basic RDF. Full expressiveness requires OWL-DL or
OWL Full. The use of RDFa and simple RDF vocabularies like FOAF is not
a strong example of how semantics can be applied.

Bill again pointed the TC to the Description Logic navigator at
http://www.cs.man.ac.uk/~ezolin/dl/. He made the point that the
marketability of XDI is largest if semantics is at its very core.

Markus asked about an example. Bill gave the example of describing a
social network. He cited a recent analyst prediction by that by 2016
20% of our friends in our social networks will be automated agents
that process the information we receive from our social contacts. He
also explained the value of goal-oriented reasoning in this scenario.


3) REVIEW OF XDI GRAPH PATTERNS

We reviewed the different patterns on new wiki page Drummond posted last night:

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

# ACTION: Drummond to complete these diagrams for link contracts and
dictionaries.


3) NEXT CALL

We did not have time to review the December call schedule. Drummond
may have a conflict next week but he will do his best to work around
it.


------------
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