[DAIS-WG] Minutes from the DAIS Call on 7th June 2010

[Steven Lynden]

DAIS Call - 7th June 2010
===============

Present
-------
    Mario Antonioletti, EPCC
    Mike Jackson, EPCC
    Oscar Corcho, UPM
    Carlos Buil Aranda, UPM
    Isao Kojima, AIST
    Steven Lynden, AIST

Agenda
------

- Review of actions from the last call.
- WS-DAI/WS-DAIR progress, including recent changes and proposals
about the DatasetTooLarge fault.
- Recent work on the RDF implementations.
- Planning.

Actions
-------

[MA] Chat with the WS-DAI/WS-DAIR specification reviewer and try to
get an estimate of how long it will take before we can get the review.

[SL] Update the RDF querying document & WSDL to reflect recent changes
to WS-DAI and WS-DAIR.

[OC] Inform Joel Replogle that there will be no DAIS session at the coming OFG.

[SL] Send a final-call email to the DAIS-WG list asking for input on
WS-DAI/WS-DAIR before submission (when we have some feedback regarding
the review).

Completed actions
-------

[MJ] Do a final check of the changes made to the WS-DAI and WS-DAIR documents.

[SL] Send an email to the WS-DAI/WS-DAIR specification reviewer to
find out about the current status of his review of the specifications.

[SL] Compile a list of people/projects using the WS-DAI family of
specifications and put this info up on GridForge.

--

The group discussed progress with WS-DAI/WS-DAIR.

There is concern that the review is taking a long time. The feeling is
that we should go ahead and update the RDF querying specification to
reflect the WS-DAI/WS-DAIR changes now instead of waiting any longer.

Pressing ahead with the RDF work is also important as Carlos needs to
wrap up his work on this by 20th July.

Carlos gave an overview of the current status of his implementation.
Everything is going well, the service interfaces have been created
without any problems using the latest WSDL and deployed using a
Glassfish server. He'll use Jena to provide the RDF DB.

AIST can still do interoperability testing after the 20th July as long
as the UPM implementation is working. Steven will update the RDF spec,
then Carlos and Steven will start an interoperability testing document
and get as far as they can.

Interoperability testing will be completed at AIST as 20th July is too
ambitious a time-frame for this.

Following on from recent messages on the mailing list,
DatasetTooLargeFault was discussed. Consensus was reached and the we
decided to go back to the previous proposal of just having the fault
for direct access operations when the dataset is too large for direct
access only, thereby suggesting the use of indirect access to the
client. Having the fault thrown for factory operations seems to go
out-of-scope and its not exactly clear under which circumstances it
should be generated.

DONM
----

21st June - 7am GMT/UTC