[INFOD-WG] [Fwd: Info-D M.Sc. project]

[Dieter Gawlick]

FYI

-------- Original Message --------
Subject: 	Info-D M.Sc. project
Date: 	Fri, 19 Jan 2007 17:55:48 -0000
From: 	Dave Berry <daveb at nesc.ac.uk>
To: 	<dieter.gawlick at oracle.com>, <S.M.Fisher at rl.ac.uk>, 
<vdialani at us.ibm.com>, <ronny.fehling at oracle.com>



Dear Dieter, Steve, Vijay and Ronny,
 
Stephen Davey's parting gift to me on leaving NeSC was a proposal for an 
Informatics M.Sc. project to implement part of the InfoD specification.  
I'm happy to say that this proposal was approved and I now have a 
student assigned to it.  He will spend 3 months this summer working on 
the project.
 
In the meantime, he and I have to flesh out the project plan.  I'm 
hoping that we will be able to ask you for advice on key points, as 
neither of us have any prior knowledge about the details of the InfoD 
specification.
 
As this is an academic project and not just an implementation exercise, 
a key part of the project will be working out how to evaluate the 
system.   I would value your suggestions on this point.  It seems to me 
that while we could evaluate performance relative to a.n.other messaging 
system, the actual transfer of messages is not the focus of InfoD, so 
all we would hope for is that our performance would not be degraded by 
comparison.  My understanding is that InfoD is more about improved 
flexibility, so it would be worth demonstrating some examples that could 
not easily be implemented using an alternative system.  Do you have such 
examples already?
 
I would also appreciate pointers to alternative messaging systems with 
which to compare the new system.  Would the Globus implementation of 
WS-Notification be a useful comparator?  Or should we be looking to 
build our InfoD implementation on top of WS-N?
 
Stephen and I assumed that we will not be able to implement the full 
InfoD matching algorithm in a 3-month project.  We suggested to start 
with a simple messaging system and a basic registry.  Once the student 
has completed that infrastructure, he can start to add some basic 
constraints and investigate constraint solving algorithms.
 
As a background question, do you have a formal treatment of the InfoD 
constraint system.  E.g. a specification of the constraint language in 
formal logic with an associated decision procedure?
 
Best wishes,
 
Dave Berry
Deputy Director, Research & E-infrastructure Development
National e-Science Centre, 15 South College Street
Edinburgh, EH8 9AA                  +44 131 651 4039
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.ogf.org/pipermail/infod-wg/attachments/20070124/5b4a94f3/attachment.htm