Eric J Ma's Website

PyCon 2017 Highlights

written by Eric J. Ma on 2017-05-22 | tags: pycon conferences python

Last post was about thoughts on past PyCons, having attended PyCon 2017. This post is on PyCon 2017's highlights for me.

(1) Serving as part of the organizing committee. I had the privilege of serving on the FinAid committee this year, and spent a large fraction of time in the staff room preparing to disburse FinAid cheques. I have very vivid memories of how slow the line was when I was receiving my cheques back in the day, and so I wanted to make sure FinAid recipients could receive their reimbursements as fast as possible, without wasting time in line (when they could instead be listening on talks).

(2) Teaching two tutorials. This year, I submitted two tutorial proposals, and both were accepted. In the three years that I've been teaching it, Network Analysis Made Simple has always been popular, and I think it's because it gives participants a different way of thinking about data, thus making it an intellectually stimulating topic. I also developed a new material on Best Testing Practices for Data Science. This one, in retrospect, was much fresher, and thus in need of more battle-testing and polish compared to Network Analysis. I have some ideas, including modifications to the workshop format, narrowing the target audience and more, to make it more useful for future iterations.

(3) First talk at PyCon! I also gave a talk at PyCon on doing Bayesian Statistical Analysis with PyMC3! This was my first PyCon talk ever. It was so nice to have a tweet-commendation by PyMC3's creator Chris Fonnesbeck too:

It was also nice to have Thomas Wiecki's tweet-commendation too:

Beyond that, the attendees seemed to like the talk too on the Twitterverse!

It's very heartening to see how many people want to move into Bayes-land! The talk also happened to be the last in the session and last of the day, so I think many people were tired by that point and wanted to go to the final keynote. Thus, the only question came from my friend Hugo, with whom I also worked on a course at DataCamp, who asked about "how we might communicate these ideas to, say, a manager." My thoughts on that were to report not a single number (e.g. the mean), but also the range, and communicate how the lower and upper bound of the range would affect bottomline decisions, or open up new opportunities (though I probably could have expressed this sentiment better).

(4) Feeding Guido van Rossum. Python's BDFL, Guido van Rossum, wandered into the staff office asking to see whether the speaker ready room was open, because he was hungry and was looking for some snacks. We initially suggested the main conference hall, but later I ran out and called him back, because we had some English biscuits in the staff room, and we engaged in a short chat. That's when I had my star-awed moment! Was tempted to get a photo, but I figured he'd probably be fed up with people asking for photo ops, so I decided against it, hoping to be considerate for him. When he finished the biscuit, he said goodbye, and left the staff office. Amazing how everybody else just went about their own business while he was in the room; speaks to the lack of ego that PyCon celebrities have, and that sets a great example for the rest of the community!

Once I'm back in Boston, I'm definitely going to catch up on the rest of PyCon. I heard that there were a lot of good talks that I missed while staffing the conference as FinAid co-chair, will have to make sure that YouTube playlist is set up!

I send out a newsletter with tips and tools for data scientists. Come check it out at Substack.

If you would like to sponsor the coffee that goes into making my posts, please consider GitHub Sponsors!

Finally, I do free 30-minute GenAI strategy calls for organizations who are seeking guidance on how to best leverage this technology. Consider booking a call on Calendly if you're interested!