Minutes: XDI TC Telecon Thursday 1-2PM PT 2008-03-20

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

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

Date:  Thursday, 20 March 2008 USA
Time:  1:00PM - 2:00PM Pacific Time

ATTENDING

Kermit Snelson 
Markus Sabadello 
Drummond Reed 
Bill Barnhill
Giovanni Bartolomeo 

REGRETS

John Bradley 


1) CALL TIMING

We discussed the Daylight Savings Time issue. It turns out that Italy will
also be going on DST next week, which puts the time gap between Tokyo and
Rome so far apart that there is no time that works for both of them and the
USA.

So for the time being we will maintain the current call time (1PM PT/4PM
ET).

# DRUMMOND to send Nat an email inquiring about alternatives, including
adding a USA/Japan call.


2) CHARTER CLARIFICATION

We reviewed the XDI charter clarification page and filled in the rest of the
proposed deliverable dates:

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

The only action item left is to schedule it for a ballot.

# DRUMMOND to send a separate email to the list notifying members of a
ballot after next week's telecon.


3) CO-CHAIR TRANSITION

Geoffrey Strongin is ready to pass on his co-chair role to another active
member of the TC, particularly as we move into the active spec drafting
cycle.

Kermit nominated Bill, and Bill accepted the nomination.

# DRUMMOND to send a separate email to the list asking for other nominations
over the next week, and proposing a vote after next week's telecon.


4) XDI RDF SUBJECTS DISCUSSION

RE the active thread on the topic of XDI RDF subjects, Bill sent an email to
the list explaining that the W3C RDF Core Working Group discussed allowing
literals as subjects and in fact decided in favor of it, but did not act on
it as that time.

	http://lists.oasis-open.org/archives/xdi/200803/msg00043.html 

Bill clarified that he does not have a compelling use case for doing this.
He believes the more important requirement is that every subject MUST be an
addressable XRI to maintain the addressability of the XDI RDF graph.

There was consensus on this point. Therefore the decision was that a literal
must be transformed into a relative XRI in order to use it as an XDI RDF
subject.

Drummond also pointed out that he believes the "$ pattern" that emerged in
the discussion of this subject is a very useful pattern that will apply in
many places besides XDI RDF emulation of HTML links. See:

	http://lists.oasis-open.org/archives/xdi/200803/msg00038.html


5) REFLEXIVE PREDICATES

We also discussed the other part of that thread on reflexive predicates,
specifically $is. The conclusions from this discussion were:

* $is and $equals are different predicates in that $is asserts synonymity
between two XRIs (that they represent the same resource), while $equals
asserts that the _literal values_ of two XDI statements are equivalent.
Example:

	=a/$is/=b
	(=a/+age)/$equals/(=b/+age)

* While $is is a reflexive predicate, i.e., =a/$is/=b implies that
=b/$is/=a, these two XDI statements still represent _two different arcs in
the XDI RDF graph_. Each is the backpointer to the other, but they are
separate, and one can be verified by doing a query on the other (i.e., if =a
asserts that =a/$is/=b, then =b can be queried for =b/$is=a to see of =b
agrees). Note that this is identical to verifying EquivID synonym assertions
in XRI Resolution 2.0.


5) DOLLAR WORDS

Giovanni talked us through the key parts of his proposal at:
	
http://www.oasis-open.org/committees/download.php/27547/Draft-Contrib-OASIS-
XDI-dollar-words-v03.pdf

It included a new $ word, $any, for use in queries. This can be used to
match against any XRI or XRI subsegment. A detailed example is in chapter 9
of his document. This has been further discussed on the mailing list to
optimized the query syntax for using $any.

The proposal also includes $all. Whereas $any is like a Prolog variable --
it sets a variable instance, $all is a true wildcard.

On the email list, Markus has suggested $any$1, $any$2, however this could
be ambiguous as to whether the $x represented an instance of the $any
variable, or a query to be matched. One suggestion to solve this was to
define $$ as the $ word for an XDI variable.

Giovanni's proposal also includes $or and $and as query operators, shown in
chapter 7. This led Bill to ask if Giovanni had proposed $not or $xor. 

Giovanni noted that $not is a unary predicate, so this constrains the way in
which it can be used. It could be used to negate the statement that follows.

Bill suggested that Giovanni's document be migrated to the XDI TC wiki so
that we could start to build the $ word operators incrementally. Drummond
agreed and proposed using the current DollarWord page as a master index page
for all $ words and $ word proposals:

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

# GIOVANNI will start converting his document to pages on the wiki, with
links to these pages from the DollarWords page above.


6) OASIS SYMPOSIUM DINNER, DATA SHARING SUMMIT, AND F2F MEETING

Drummond provides a reminder that we have three meeting options coming up,
all in the San Francisco Bay Area:

	- Dinner during the ORMS TC Meeting on Thursday night, May 1.
	- BOF during the Data Sharing Summit Part A April 18-19
	- F2F during and the day after the Data Sharing Summit Part B May 15

See:
	http://www.datasharingsummit.com  

Bill will only be able to make the first of these. Drummond plans to attend
all (although only the first day of April 18/19 Data Sharing Summit). We'll
do more detailed planning as these dates get closer.