Minutes for LArSoft coordination meeting on July 17, 2018

1. Simulating S2 light in LArSoft 30'- José Soto
   Since the S2 light is produced by electrons after they drift to the anode region, the design reveals a new interface layer that carries information about the drifted electrons in a data product, and that can be used by an abstract anode region simulation.
   Breaking change in larana — the input module argument is now a vector, and a loop introduced to go over elements of the vector. Herb noted that this could be made to accept either a string or a vector as input to the fhicl Making it non-breaking. Herb and Paul have the details.
   Approve putting in
   
2. Restructuring anab::ParticleID — Adam Lister
   While doing PID work on MicroBooNE, they developed a flexible reorganisation of the anab::ParticleID class that allows for new PID algorithms with arbitrary output variables, and makes the class useful for physics. The main downside is that this relies on the analyser knowing what’s in the structure, and so it requires good experiment-specific documentation. The changes were approved for release after some minor modifications are completed.
   Accommodate direction somehow? Add a new enum for the direction. Have legacy methods access new structures, or as was discussed after the meeting, we determined that the approved approach would be to remove the legacy methods.
   
3. LArSystematics: systematic shift framework for LarSoft — Luke Pickering
   The code is derived from experience from MINERvA and T2K, and NOvA in building/adapthing neutrino interaction systematic propagation software for use in DUNE TDR sensitivity studies. DUNE intends to use the code in both the near detector analysis (Currently EDepSim) and the far detector analysis (LArSoft). It is general, so an appropriate candidate for shared analysis software in LArSoft.