Agenda

  1. Fix to LCIO                                              — Norman
  2. Organizing the calibrations                      — ALL
  3. Data Blinding
  4. Software Roundtable                               -- All
  5. Questions?                                               -- All

Notes

On Data Blinding: We decided that a reasonable approach was to process only 10% of the raw EVIO files with reconstruction. We would do every 10th file, so hps_xxxx.evio.0, hps_xxx.evio.10, etc. This should give a good representative sample of the data. The overall data set is large enough that it is highly unlikely anyone could process it and not get noticed. We briefly addressed the potential issues that may arise when it is time to un-blind, since it is likely that one person is further along with the analysis and thus ready for un-blinding, when others are still working on theirs. The EC will make the decision on when to unblind and how to restrict access to the full data set to only those analyzers that are ready for it.

 

  • No labels