Tamara has used her cluster density weighting scheme to identify different types of clusters:
A large ratio of energy/cell-volume indicates a more em-like cluster while a low energy/cell-volume indicates a more hadronic-like cluster. Taking into account the cluster type, she arrives at a better resolution, and at a response much closer to 1 than before (without the density weighting). The response is also more linear, a sign that software compensation is working.
Tamara will start writing her thesis next week.

Warren tried to get Sven's and Pavol's code to run. He finds that it crashes after approximately 300 events. It does not always crash at exactly the same event though, which indicates, that there is a problem with the code.
Warren also started to look into how to do Python scripting.

Margret and Ian have come to the conclusion that the LArG4TBEmecHec package code will need a thorough re-write in order to implement a Calibration Ntuple in a clean way. They will start by making an outline of what exactly needs to be changed and which parts of the code are no longer required. The final result should be a new package (LArG4TBEmecHec02) that no longer contains relics of the Geant3 HEC version and will have namespaces that are analogous to the LArG4 code.