-----
Heiko Ludwig, Dr. rer. pol.
IBM TJ Watson Research Center, PO Box 704, Yorktown, NY, 10598
hludwig@us.ibm.com, tel. +1 914 784 7160,  mob. +1 646 236 9453
http://www.research.ibm.com/people/h/hludwig/
----- Forwarded by Heiko Ludwig/Watson/IBM on 04/20/2005 11:04 AM -----
Heiko Ludwig/Watson/IBM

04/20/2005 10:13 AM

To
Jim Pruyne <pruyne@hpl.hp.com>
cc
Asit Dan/Watson/IBM, John Rofrano/Thornwood/IBM
Subject
Re: [graap-wg] minutes from telecon on 4/13/05Link




Jim,

I updated the draft as per our discussion last week. I didn't find a proper way to add a document to the GridForge that supercedes the current draft. Hence, I send it to you for you have the authorization as an admin. Since the call is near, we also might just want to send it to the mailing list.

Heiko



-----
Heiko Ludwig, Dr. rer. pol.
IBM TJ Watson Research Center, PO Box 704, Yorktown, NY, 10598
hludwig@us.ibm.com, tel. +1 914 784 7160,  mob. +1 646 236 9453
http://www.research.ibm.com/people/h/hludwig/


Jim Pruyne <pruyne@hpl.hp.com>
Sent by: owner-graap-wg@ggf.org

04/13/2005 12:21 PM

To
GRAAP-WG <graap-wg@gridforum.org>
cc
Subject
[graap-wg] minutes from telecon on 4/13/05





They are attached.  We'll continue the current meeting times.

--- Jim

                                    Minutes from Apr. 13 '05 Telecon


Attendees
---------

Jim Pruyne
Heiko Ludwig
Karl Czajkowski
Wolfgang Zeigler
Asit Dan


Agenda
------

- What's up with the Author's list?  Jim's not comfortable removing
 names without their consent, and he hasn't gotten consent from
 people when he's asked.

- Summary of updates from Heiko's revision
 - Change to Penalty to contain an unbounded list of (Assesment
 Interval, value unit, value expr.) tuples.  Proposed to just change
 so that we can have an unbounded number of Penalty clauses in the
 BusinessValueList  **Action: Do this in the spec.  Heiko will do
 this, and apply to Reward, etc.
 - Discussion of comment 29: Are the value terms really appropriate
 here, or does this need to be addressed someplace else? **Action:
 Response is that there is nothing else today, so this is an approach
 we suggest.  This specification can considered optional in the sense that
 one need not include guarantee terms at all, but is useful in some
 domains we've looked at.
 In 4.2.5.1, we need to
 be clear on what parts are extensible to support this position.
   - Does this all lead to the need for extensible term types?
   E.g. an "option" as suggested by Heiko.  It seems that there may
   be terms beyond the description (functional) and guarantee
   (non-functional) terms that we have now.  Agreed that we should
   allow for extension of new term types.  **Action: update section
   4.2 to make the definition of new term types possible, and update
   the pseudo-schema in 4.2.1, and introduce a new-subsection before
   the compositor definition.
   *Comment: Use of xsd:anyother is a useful construct to specify
   that one can extend, but must use another namespace outside our
   own.
   *Comment: How do we make it clear in the pseudo-schema where
   extensibility is possible, and can we summarize somewhere all of
   the extensibility points.  
   *Comment: Should we try to do some re-structuring to avoid the
   high-level of nesting that we have now (e.g. 5-levels of section
   numbering)

**Action: Karl to do a re-write on 2.1 on motivating example on Job
Submission, but it needs to stay up-to-date wrt later sections and
examples in appendicies.