== Found Problems ==

 - In the input events we have particles coming from proton remnants and Ks decays. These decays can happen quite far from the interaction point. When the particles produced far from the interaction point are propagated to the calorimeter, their momentum eta and their position eta are quite different. What should we do with these decays? Should the different types of particles (charged, neutrals, electrons, photons, muons etc) be treated differently?

 - Calorimeter towers are required to have total energy greater than zero (Eecal + Ehcal > 0). What should we do with towers that have negative Eecal < 0 or Ehcal < 0? '''Solved: set negative Eecal or Ehcal to 0.'''

 - Only calorimeter resolution is used for electrons. Recent experiments often use tracker system to improve momentum resolution. Should we try to simulate these effects?
 
 - Towers that have at least one hit from a neutral particle are used together with tracks as input for the jet finding algorithm. If a tower has hits from both neutral and charged particles, then we have double counting for the charged particles. '''Solved: calculate fraction of energy from neutral particles in ecal and hcal and then apply this fraction to the smeared ecal and hcal energy.'''

 - Current tau-jet finding algorithm seems to be too simplistic:
   - E_cal = sum of energies of calorimeter towers with ET_tower > 2GeV within DeltaR < 0.15 around jet axis
   - N_trk = number of tracks with PT_trk > 1 GeV within DeltaR < 0.4 around jet axis
   - select jet as tau-jet if PT_jet > 10GeV and E_cal/E_jet > 0.95 and (N_trk = 1 or N_trk = 3)

 It looks like a simplified version of the tau-jet finding algorithm from PGS:

 http://indico.cern.ch/getFile.py/access?contribId=1&resId=0&materialId=slides&confId=187127

 And the latter is reported to have problems:

 http://indico.cern.ch/getFile.py/access?subContId=5&contribId=6&resId=0&materialId=slides&confId=187127

 Should not we implement tau-jet finding in the same way as we implement b-jet finding?