Meeting to discuss paper for e-science all-hands conference
===========================================================


Mark will be mostly talking about gridmon and where it's going.
But it won't be going towards the justification of our stuff either.


Yee is away on the last day of the meeting definitely.
Apparently the last day is usually just a few keynote speeches.


Audience: very mixed: people who know about network monitoring, and people who are jsut site admins, say.

20 minute slot, 5 mins for question. So, after 5 mins of introduction there's only about ten minutes of technical detail.
Which just about shows how much detail you can get into.
The idea is not to impart lots of information, but to give people bite-sized chunks of information about what sort of stuff is going on.

We were going to focus primarily on the domain interface design.
Mark: most people won't be familiar with pipes or dante, so we can start by going into that.
The basic layout, how they are interoperable, that sort of thing.
Then say finally exactly what we're doing at UCL.


Highlight the fact that it's a very international colaboration. Shows it's good value for money, therefore.



People to write bits:

Yee: architectures
Mark: description of pipes and dante; why what we're doing is really good
Paul: Technical side of domain interface