Marlon:
I am interested in contributing/leading Section 2: Metadata representation; but not do it without at least one other person (several would be better).
There are lots of examples in various sciences (but the grid is not always popular) where metadata, schema mediation, workflow composition, etc.. are needed. Personally, I can contribute stories for climate sciences and nano sciences (DOE SNS, new user facility).
Many of the physical sciences are only starting to recognize their needs which vary quite a lot from one to another. For instance:
*small user community, large data sizes, multiple files and archives. (example: climate)
*hundreds of scientists but small files available on personal machines (not archives). (example: bio). However the size of files is changing fast. There is also a need to discover networks of interaction between variables working at various scales.
*flexible workflows: bio
*fixed workflows during an experiment, the course of a simulation, and/or in the way of doing a particular science (fusion, combustion). Here workflows tend to be hardwired.
*cultural issues: some sciences are more ready to accept new technologies than others, many have been burned in the past, others plainly don't have time.
The primer can do a lot to dispel some myths here (but not expect miracles!). The examples will need to be picked out carefully in order to really highlight where the Semantic Grid can make a difference, and also address a community relatively open to new technologies. In other words, the killer app.
Line
-----Original Message-----
From: owner-sem-grd@gridforum.org [mailto:owner-sem-grd@gridforum.org] On Behalf Of Marlon Pierce
Sent: Wednesday, June 09, 2004 1:43 AM
To: sem-grd@gridforum.org
Subject: [sem-grd] Primer meeting
I've attached (again) the proposed prime outline. We will be meeting to
work on this tomorrow at 1:00 pm in South Pacific Room 3. Note the
primer itself is intended to contain shortened versions of #2-7. After
a discussion with Jim Myers, I'd like to emphasize that we have two
intended audiences: GGF types and application user types. For the
latter group, we really need some short but intriguing examples of what
the Semantic Grid should enable for various fields: bioinformatics,
chemistry, etc.
Marlon