Ashok reported that has brought the source code from CERN to UVic (fate), and he did it by means of an ssh protocol. He was able to download everything he wanted to get, but hasn't had a chance to compile it yet. Everybody was very interested in this. It was suggested that Ashok pass on this knowledge to Ian Gable and to Warren Shaw and eventually to write down the instructions on a web-page.
Ashok told us that David Quarrie was extremely helpful with this endeavour.

Warren received an email from Rob with instructions for the monitoring. He has tried it out and it works, but hasn't received any instruction of what to do with it next.
Richard pointed out that ultimately it should be possible to run the monitoring code here at UVic, looking at data that resides at CERN, i.e. a remote monitoring of the testbeam.

Tamara presented progress of her noise studies. She looked at muon runs to obtain noise from cells that were not hit by the muon beam. In order to get a signal-free noise sample, she actually took muon runs taken at different impact points and only chose cells away from the impact points for the noise sample.She then plotted the average signal/noise ratio with the signal from pion runs and the noise from muon runs. She is now looking at what the best way might be to make a selection that removes the background due to noise.
Richard suggested to choose a typical cluster region and calculate the correlation coefficients between the cells. - Michel mentioned that something like that has already been done by Manuella, who wrote a note about it.

Tayfun explained some more insight he gained into the systematics of the position reconstruction analysis. With a simple diagram he demonstrated that as the beam moves across the front face in an x-scan, the energy sharing between the cells changes. In fact it explains some of the effect he sees, but not all of it. He recalled that Sven earlier on had come to the conclusion that the testbeam coordinate system is not perfect (it has a bit of an offset) and that the beam is likely not completely horizontal when B9 is switched off.
Richard suggested to look at the high impact points and see what effect he gets there. A small tilt in the module could produce the same effect, and it might be impossible in the end to really pinpoint the exact cause. Michel gave Tayfun some quick advice of how to go about to test for a module tilt within TBRootAna.

Michel reported that the clustering algorithm is now fully functional in TBRootAna, and that he already ran lots of tests on it. He presented some results with plots of numbers of cells per cluster and event and compared the clustering algorithm with a simple selection of taking all cells above a 2 sigma threshold. See the TBRootAna example web page.

Margret mentioned that she obtained a reply from Fares, who said that the file he sent a year ago for the boundaries between the 3 EMEC longitudinal segments are numbers for the warm module. But the differences between the warm and cold numbers are very small.