I went to the Python in Heliophysics Community (PyHC) meeting last week, and have a few updates relevant to us:
- I’ll be hosting the next PyHC meeting at the Center for Astrophysics in Cambridge, Massachusetts, USA from about April 27-29, 2020.
- Romain Dupuis gave an interesting talk on AIDA, which is an effort related to machine learning in heliophysics. It sounded like there may be some overlap with what we want to do with TurbPlasma (i.e., building computational tools to investigate turbulence, etc.).
- There was an unconference session about whether or not heliophysics needs an online science platform like Pangeo. (I missed that, unfortunately.)
- I gave a tutorial on scientific software design principles, primarily on the SOLID principles. It’s still a work in progress, but I could potentially give an updated version of that in the future for us (though probably after a tutorial on object-oriented Python). It’d also be good to be able to introduce ~3 of the most useful design patterns that come up in science (factory method, strategy method, etc.).
- Stuart Mumford gave a live tutorial on contributing to the PyHC gallery. We should really create something for that after version 0.3.0 comes out (which will be after a bunch of API changes).
- I encouraged PyHC members to propose an NSF CSSI proposal next year (like the one from PlasmaPy this past year).
With respect to meeting organization:
- Unconferences are great!
- When having hackathons at meetings, it seems like it works best to have them as the last part of the meeting.