Reconstruction, February 9, 2008

Hi Matt:

Blake and I had a chat today about things we believe should be discussed in our meeting tomorrow morning, at 11am ET. Please circulate this email and invite Ryan and Mihajlo of you think they should be there. They can also join our discussions at any other time in the future. Let's use the standard ESNET for tomorrow.

1. Blake's Progress: Code compiles ok and the 'calibration' for the 6x4 segmentation is done to the 99% level. Blake now has a stable fit function, and with your help has corrected the z position (entry versus shower center). If everything works and nothing breaks, the 3x3 can be done in 2-3 days.

2. Code understanding: Blake is reasonably familiar with the analysis code, but a lot of the HDGEANT stuff is still largely a black box. He has concerns about the non availability of 'standard releases', an issue that should be addressed at a Hall-D software meeting. Currently, upgrades/checkins cause hiccups and delays. Notes on patches should be more detailed.

3. Analysis: this is the main part of our IU-Regina collaboration. - pi0 and eta factories exist in HDGEANT. Are they ultimately the ones we want to use? Do they need tuning? - we should make a step-by-step analysis plan for Blake: which signal channels and what is important to look at plot-wise, and Blake needs to understand the importance of each step in the analysis. Also, which efficiencies to we wish to extract, chi-squared info or likelihood -- basically which quantities and metrics for comparisons. Then, what parameters should we look at: we spoke about threshold, segmentation of readout, PDE-DR info, attenuation length of fibers... - signal channels versus Pythia background. How should this proceed? - long term: PWA. Are all the necessary tools ready? They must be for Blake to use them in his tight time frame as there is not that much time for significant development.

We should discuss the timetable of tasks and come up with realistic milestones. Following progress on the reconstruction, we can then plan a trip to IU for closer interaction. Until that time, how do ESNET videoconferences every two weeks sound?

Cheers, Zisis...