Some mostly-fairly-minor comments on the code in meas_astrom at https://github.com/lsst/meas_astrom/pull/106. In addition to those, it would be great to clean up the history by squashing — per my confusion above, the unsquashed history makes it hard to understand what's going on.

Aside from that, the changes in obs_cfht and obs_subaru look fine.

Two naive questions:

• Why aren't changes necessary in other packages? It looks as though Jointcal, pipe_tasks and a few other packages are referring to OptimisticB only config parameters — why don't they need updating? Am I missing something obvious?
• I know you've demonstrated that PPMb performs well in earlier tests, but do we know exactly what impact merging this will have on the existing metrics that are being tracked in SQuaSH?

Yusra AlSayyad, Jim Bosch — do you have any concerns merging this with regards to HSC processing?

No concerns from me. We'll need to remember to mention it to NAOJ - I'm sure they'll want to do some validation of their own before using it in an official HSC DR, but even if something goes wrong there we can just disable it via config.

Good to go, modulo a tiny typo (https://github.com/lsst/meas_astrom/pull/106/commits/09c3b8f1a0d28a28a8dc83fff2ac8e4c346ca930#r249313748)

Final successful Jenkis run check: https://ci.lsst.codes/blue/organizations/jenkins/stack-os-matrix/detail/stack-os-matrix/29300/pipeline/46