I'm checking in with a post-workshop report on lessons learned from hosting my first solo astropy workshop at (my new home institution) the University of Bern, Switzerland. Workshop materials are here for posterity.
Responses from the intake survey are attached and are available as a spreadsheet here, which I created specifically for this workshop. Main takeaways:
- 100% of attendees said this was their first astropy workshop
- 90% of attendees said they use Python in their research, but to my surprise...
- 40% of attendees said they've never used astropy before
I think this result is neat, it shows how much we have to gain by going around and giving these workshops. There are still plenty of untapped users out there!
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 from my home-spun curriculum on object-oriented programming, HDF5 in Python, and releasing your own python package using the astropy package-template. The package-template demo got some audible gasps.
Responses from the outgoing survey are attached and are available as a spreadsheet here. It looks a lot like the outgoing survey in the 2019 AAS astropy workshop. Main takeaways:
- The section people perceived as least clear was the section on astropy.coordinates. I think this is an artifact of the ~70% of attendees who consider themselves theorists, who said they don’t work with coordinates regularly.
The schedule of the workshop looked like this, and the timing was mostly on target. The one section that took significantly longer than the allotted time was the Intro to Python, which took an extra 20 minutes. If I recall correctly, the same was true at the 2019 AAS workshop – I think we might need to allocate more time (one full hour?) for that section. I also cut the coffee breaks down from 30 minutes to 15 minutes, because the 30 minute breaks became “Q&A with Brett over coffee” sessions, so I cut each one a bit short so we could do the same Q&A at our laptops.
We should probably write up docs to standardize this workshop-hosting process for future trainers. Here are some tips that helped make this workshop run smoothly:
- Allocate at least a full workday (eight hours) to prep for the workshop – testing the full workshop takes longer than you’d think
- Have a dedicated “setup hour” before the workshop where people can come in and debug their installations before the workshop begins – we had an extra help session at 9am and had zero (!) users with installation bugs when we began the tutorials at 10 am
- Arrange for extra power chords to be available in the room – you can never have too many for a day-long workshop
- Be sure to have green+red stickies for help flags – people are more likely to ask for help when they have a protocol
- Provide coffee and snacks – people come back from breaks recharged and ready to go
- Always make the text larger on your screen – accessibility matters
- Have experienced helpers who float around during the exercises – helpers fix problems as they happen, minimizing user downtime
- Send emails about the workshop one month before, one week before, and the day of the workshop – advertising works
If you like, I can make a PR to the astropy-workshop repo (or somewhere else?) with a checklist with these items.
Thanks again to Kelle for the astropy stickers, they were a big hit! Based on my careful observations, 25% of attendees stuck their stickers on their laptops by the end of the workshop.