
2006.12.26

 work still to be done for psastro:

 - test suite
   - create a DVO database from a fake population
   - create fake images with range of errors drawn from DVO

 - 


we have a few different astrometry circumstances for our complete
image/chip hierarchy structure:

- ChipAstrom with one readout, one cell per chip:

  whether or not there is more than one chip, we need to have two
  stages
  
    - in the first stage, the per-chip results are applied to the
      chip.toFPA parameter set.  

    - in the second stage, the chip results are collectively used to
      modify the fpa.toSky terms.

      - the average offsets of the chip positions relative to starting
	coordinates are used to calculate a single average offset to
	the boresite. 

      - the average rotations of the chips relative to their starting
        rotations are used to calculate a single average rotation of
        the boresite


- ChipAstrom with more than one readout and/or cell:

  - match all stars up relative to first readout?
   
