I’m checking in again with a post-workshop report from hosting my second solo astropy workshop at Geneva Observatory on Nov 14, 2019. Workshop materials are available here.
Responses from the intake survey are available as a spreadsheet here. Main takeaways:
- 97% all participants use Python for their research, yet...
- 25% of participants have never used astropy
I think this echoes my result from the Bern workshop – it shows again how much we have to gain by giving these workshops.
Like in Bern, I tailored the curriculum to people’s requests on the intake survey to minimize the sense of overwhelm felt at the 2019 AAS workshop. The attendees voted to get extra lessons on specutils, plus my home-spun curriculum on object-oriented programming and releasing your own python package using the astropy package-template. Several people came up after the workshop to thank me for the packaging tutorial and told me they intend to use the package-template to release their own packages. I’m continuing to think of ways to standardize and distribute the packaging tutorial that I’ve been doing, and currently drawing a blank. If someone has a good idea for how to make a tutorial that walks you through the interactive cookiecutter dialog, I’d be keen to know it!
Outgoing survey results are here. Main takeaways: – The sections people perceived as least clear were FITS and Tables. It seems that these respondents who found the sections least clear wrote that they thought the sections went by too fast. This is likely an artifact of the fact that I assumed most of the users have seen the FITS module before, since most attendees had said “our only experience with astropy is io.fits”, and >70% of respondents were observers. – A few respondents suggested that we give them introductory materials in advance of the workshop.
The schedule of the workshop looked like this, and the timing was mostly on target. Again, the one section that took significantly longer than the allotted time was the Intro to Python, which took an extra 10 minutes. I think this was largely due to the fact that many of the users were inexperienced with Python and didn’t do any preparation for the workshop so we had many questions. I think this echoes my earlier suggestion that we might need to allocate more time (one full hour?) for that section and we need to send out “warm-up” material, like was suggested in #60.
We only had a few attendees at the “setup-hour” from 9-10:00 before the workshop began, but remarkably there were zero installation/environment problems when we began promptly at 10. It looks to me like the instructions in the repo are really well battle-tested at this point.