** Participants **
Chris Smith, Platform
David Snelling, Fujitsu
Michel Drescher, Fujitsu
Andreas Savva, Fujitsu
Pete Ziu, Northrop Grumman
Glenn Wasson, UvA
Marvin Theimer, Microsoft
Hiro Kishimoto, Fujitsu
Nick Werstiuk, Platform
Ian Foster, ANL
Jay Unger, IBM
Andrew Grimshaw, UvA
Ravi Subramaniam, Intel
Steven McGough, Imperial College
Neil Chue Hong, EPCC (via Phone)
Dave Berry, NeSC (via phone)

- Dave Berry is giving a presentation on the progress in the OGSA-DATA group

- The group discussed issues arising from slide 6. [bad phone link makes it 
  very difficult to understand Dave.]

- Ravi is concerned about an overload of terms with respect to service and 
  resource [slide11]. Ravi revokes the question due to unfamiliarity with the
  OGSA Glossary

- Hiro: Storage management has a control interface - does it also also has a
  transfer interface?
  [Answer not understood because of bad phone line quality]

- Dave Berry clarified the relationships between storage and data:
  * Storage stores data, hence two interfaces exist, for storage and data, 
    respectively [slide 12]

- Ravi asks for clarification whether Federation and Cache are specialisations
  of Data Services
  Answer: Yes in the sense that the cache provides the same interface (among 
  	  others) as a normal data service. [slide 13]

- Ravi is concerned whether the Scenario 2/5 [slide 16] is actually a cache 
  scenario

- Hiro: [slide16] why is the Transfer Service not an Access Service?
  Answer Dave B.: The scenario implies that the user reads the contents of the
  	      	  file, so she rather transfers than accesses the file (even 
		  though access is an applicable term)

- Ravi, Andrew [slide 20]: These are actually not policies by themselves but 
  	       	      	   probably parameters to a policy, or QoS related 
			   terms in a SLA

- Andrew: [slide21] Is the ParentDataResource rather a ParentResource in terms 
  	  of derived data sources?
  * Naming should be changed to "SourceDataResource"
  * These terms are not exhaustive and not exclusive enough

- Andrew: [slide21] Shouldn't this be EPRs instead of URIs?
  * According to WS-Naming an abstract name is a URI

- Andrew: DataResourceAbstractName, is this what is in WS-Naming now called 
  EndpointIdentifier?
  * The DAI properties have been specified before the WS-Naming spec, so yes
    they are.