OGSA Teleconference - 1 June 2005 ================================= * Participants Mike Berhens (R2AD, LLC) Andrew Grimshaw (UVa) Hiro Kishimoto (Fujitsu) Fred Maciel (Hitachi) Tom Maguire (IBM) Mark Morgan (UVa) Takuya Mori (NEC, ANL) Kazushige Saga (NAREGI) Andreas Savva (Fujitsu) Frank Siebenlist (ANL) Latha Srinivasan (HP) Jem Treadwell (HP) Minutes: Andreas Savva * OGSA chair nomination Tom is waiting for approval from management. No problems so far. * Roadmap document update Submit any new text to Jem using the latest draft to simplify merging. * May F2F (London) review It was very good meeting, not just for OGSA-WG but also for BES-WG, DATA-WG, ByteIO-WG. ** Minutes review A number of corrections to the minutes were submitted from Hiro. Minutes approved with the following modifications. ** Day 1 minutes - "Application manager" should probably be "Job manager." But the minutes are correct since the term "Application Manager" was used in the meeting. Agreed to add a note for completeness. - Profiles are from WS-I not W3C - SMIS should be SNIA - WSDM refers to two different WS-Addressing versions - There are two basic profile definition sections. The first should be 'profile definition' instead. - Question on plan to update BP 1.0? Not sure of context. Delete from minutes. - Consensus was to take RR out of BP 1.0. - [There is followup from Frank on list; discussed as a separate issue under "Resilient References" below.] ** Day 2 minutes - One action is listed twice. Delete the duplicate. - NaReGi should be NAREGI - Timing for RNS document handover; the minutes are correct but there are a number of problems discussed as followup issues under "Naming charter" below. * Naming charter - There seems to be some discrepancy between what was discussed at the F2F and the GFS-WG's thinking about the RNS specification. - GFS believes that they should followup through public comment and publication to GWD-P-R. - Therefore the Naming WG proposed charter would have to be revised. (Steve Crumb has not received the Naming draft charter yet.) - There is probably some confusion here. Agreed that the interested parties should arrange a separate call to clear things up. Action: Andrew to arrange a call with Hiro, Osamu, Manuel. Action: Andrew to send draft charter once the RNS document issue is cleared up. * Resilient References (RR) The main concern with taking the resilient references out of the WSRF Basic Profile and into the Naming Profile is the timing of the release of the naming profile. There is strong desire to have this functionality available sooner rather than later. Also, as long as abstract names are not required by the rebinding protocol then in the simplest case it is "rebind an EPR"; and should not require this function to be tied to the naming specification. There was an assertion that if all the information for the rebinding is put in the EPR then there is only one way of doing it. The opposite argument was also made that there are many different ways and do not want to over-constrain the choice by putting this function in the WSRF BP. Making an early choice on this in the WSRF BP would might cause OGSA to end up having multiple, different, ways of doing the same thing. (There was also an argument for taking things slowly on this in order to get the involvement of MS. There is an ongoing effort to get their buy-in.) It was agreed that it is unlikely that a resolution could be reached during this call. - A separate call or discussion on the list was proposed. - Logistically getting the right people on a call within the next week seemed impossible. - This is a simple issue and it was proposed not to repeat the discussion and vote on it instead. - It was agreed that GGF does not allow voting at this point and that discussion should continue on the list; and possibly on a call. * Teleconference schedule review June 15 call: Post-ggf14 discussion - Mark can't make the call on June 13 - Andrew will deliver Mark's ByteIO status report - Andrew can't make any calls during June 20-21 Agreed that the schedule is ok as is. * Profile Definition document review ** p3 (section 3) - Appendices as normative extensions comment - There is a discussion of this in section 3.4. - Reviewed section and agreed it is appropriate. ** p4 (section 3) - Made some clarification on Status types. In particular the Evolving standard was renamed to "Evolving *Institutional* Standard" and added the requirement that there is an external stable link available. - also "De-facto" moved to top of the list. - Added clarifying text on the meaning of adoption levels: they include the previous level, i.e., non-exclusive. - Added the EGA provisioning specification as an example. - Does this mean that EGA does standards? It is not clear since we do not have access to this specification. It may be just using existing standards. - Changed the WS-Security reference ** p5 (section 3.2.1) - Moved table from appendix and added explanatory text here - The Roadmap document should be updated with the new version of the table - Discussed how to keep tables current and provide consistent snapshots in the various documents. There was a suggestion for doing the main table in Excel and using OLE links to tables in Word. - Not sure if that would work out in all situations, for example, if there is no local copy of the master table would it have to be downloaded and so on - Entries should not be a string but a hyperlink or a cross-reference within the specification - WSDL 1.1 only added for illustration. It is not the intention to ask people to describe required specifications to this level of detail - Table needs a caption. Tom to add. - Agreed not use EGA as an example here. It gives the impression they do standards. Use some other consortium instead. Action: Hiro to find better example of consortium (done) ** p6 (Section 4) Added requirement that criteria be made at the time of submission ** p8 (References) Tom to check the OGSA Architecture, Glossary, or CMM document on how to format references in the way that the GGF Editor requires. Make citations cross-references in the text, e.g., using bookmarks Tom's revision to be picked up by Jem for further editing. * DMTF CIM - Some people would like to read up on CIM, but there are a number of different documents and are unsure of where to start. For example, is there textual description of what the meaning of the various terms are? - The MOF files describe semantics and have more information; but may not be rigorous enough - The white papers may also a good source of extra information * Next call Monday is review of Roadmap and BP