MICAC Subcommittee Meetings
Current Meeting - 17th Jan 2006
Agenda:
1) ROOT
team feedback - Sergei
2) TClonesArray
vs STL - Sue
3) How do we converge on making a decision to work on this - All
Some thoughts...
1) Manpower/Time
a) At collaboration meeting, most questions to Jim
during the reco plenary were "Will I still have the ntuples?"
b) From chats with people, it may be possible to
persuade them to take on something specific. May be contingent on the
basics of the new framework being well established.
c) Aiming for a timescale of months for completion
2) I/O speed
a) Aiming for ntuple speed, but how good is good
enough?
3) Moving Forward:
a) Should we make a new cvs branch for development
of the code?
b) How to coordinate? From the wishlist exercise,
can see there are still several design decisions to be made (event
container, etc)
c) Validation
Rustem:
I believe that changing to a new framework is not as
hard as it seems
and we should not make a decision based mostly on the difficulty of a
task.
First of all, it is almost impossible to estimate
amount of effort required without actually doing the job.
Given that, it is easier to write the code in the
new framework so I believe writing new code would be faster than last
time.
And thirdly, we have a lot more experience so we
know exactly what needs to be changed.
There are significant advantages in doing analysis
using candidates and
if IO problems could be addresses I think we should give it a try.
Minutes From Previous Meetings:
18th November 2005
6th December 2005
20th December 2005
17th January 2006
"Wishlist" Exercise
Talks
MinosSoft
2.0?
Digit
and Strip Reconstruction in the New Framework
Core
Software Parallel Session Talk at Oxford Collaboration Meeting
Reco
Session Plenary Talk at Oxford Collaboration Meeting