TODO.txt

**************
* NICOLAS: Worry about how to make cal file related routines in OneLightToolbox read/write from the right cal
  directory.  Currently the core routines respect a 'OneLightToolbox' preference, which is a good default.
  Try to make sure that all calibration related routines respect the 'OneLightToolbox' preference.
  -DONE as of 8/8/2017. Jack to test. - NPC.

* MICHAEL: Create an OLApproach_TrialSequenceMR repository and start populating it.  Or at least think about
  how the current "driver" works and what we'll need to do to make it work in the new regime.

* US: It used to be the case that blank trials were handled by a special case in make modulation starts stops
  that forced the modulations to be zero, and did a check on the direction primary cache file name to
  decide when to do this.  No No No! Going forward, this will be handled by explicitly created a zero
  contrast modulation file.

**************
COMMENTING WHERE WE NEED MANUEL TO HELP:

* DAVID, MANUEL: Add comments to: OLAnalyzeValidationReceptorIsolate and SST/ComputeAndReportContrastsFromSpds.

FOR THE FUTURE:

* There is old splatter code that we pulled out.  Probably want a way to run splatter analyses
easily, based on direction/validation files.  This should be modular.

* Organize saving and analysis of state measurements.  We make these when we calibrate, as
  we warm up the device, when we spectrum seek, and when we validate.  We should make it 
  easy to understand what we get and what we measure.
  - Also, OLAnalyzeCalOOC currently is very slow and prints out many many plots.  Most of these
    should be turned off by default.  They are useful when we are deeply trying to understand
    how the boxes work, but way too much to stare at every time.

* The direction cache files store backgrounds, differences, and +/- modulations.  This is redundant
  and can lead to inconsistencies.  Clean up someday.