(present: IG, TI, NK, AA, ML, MF, TS)

Status Reports
Tayfun has suspended his work on the beam chamber information while Naoko is working on the bug-fix for the beam chamber reconstruction. Meanwhile, he has started to look at calculations of the lateral shower shape, to get an estimate of how many cells should get hit in any compartment. Michel pointed out that MC data should be looked at for this as well. Margret offered to take Tayfun through what she has learned about the EMEC and her shower shape analysis so far.

Tamara continued her work on cell occupancy. For various thresholds (signal>t*ped_rms), she plotted the distribution of the number of cells as a function of the percentage of the time during a run that they were found to be above threshold. The aim of this effort on cell occupancy is to produce an occupancy-selected list of cells for test beam data analysis. This should be packaged in a systemAlg for all to use. Michel pointed out that he has now finished his noise-package, which should make noise estimates more reliable soon. This is likely to improve the meaning of the cell occupancy, and hence of it usefulness.

Margret reported that she has written a little cell-selection routine, that looks at occupancy within a given (fairly large) radius around the beam spot. She will pass it on to Tamara. Margret will now concentrate on transferring her analysis to TBRootAna.

Naoko showed some slides on the bug-fixes for the beam chamber reconstruction. She plotted the difference between the reconstructed position of the beam chamber fit vs that of the EMEC. She did this plot for the y-direction in the first compartment and for x and y in the second compartment. She superimposed these plots of the distributions `before' and `after' the bug-fix. An interesting feature showed up in these distributions: The y-position distribution peaks at 0.1cm for the second compartment, but for the first compartment (strips) it peaks at 4cm. It is unclear where this discrepancy comes from. Guesses ranged from a tilt of the detector to some combined effect of non-pointing geometry paired with shower development. It will have to be investigated

Ian reported that now that lots of people are actively using TBRootAna, he is spending most of his time on bug-fixes. Next big task is to improve the geometry service on nearest neighbour touching.

Michel reported that he showed plots done in TBRootAna during a phone conference earlier in the morning and said that people were very impressed with the capabilities of the package.
He also reported that his noise-estimate algorithm is essentially finished now and will be committed soon. He also finished another statistics algorithm that prints some vital run statistics at the end of each job run. He reminded us to update our TBRootAna version often, as algorithms get improved.
Another thing he did is to calculate the exact shape of the EMEC cells which are at the top edge of the detector. The corrections based on these odd shapes will be implemented soon.
Michel is now working with Ian on an Algorithm of finding touching neighbours per detector.
Finally, he mentioned that the public release of TBRootAna is still delayed a bit because of ongoing bug-fixes. But is should be coming up soon.