PyCon 2013 was my first PyCon and it was, bar none, the best conference I’ve ever been to. And it wasn’t just the free Raspberry Pi or the Wreck-it-Ralph swag from Disney or the fact that I stood next to Guido for a minute during the poster session. No, PyCon is just good people. The Python community is diverse and accepting, and I can’t list all the awesome, kind people I met there.
There were, unfortunately, disappointments, but what other tech conference has a sold-out full-day education summit, or raises $10k for a community group, or raises $6k for cancer research and the John Hunter Memorial fund with a 5k fun run? And PyCon attendees were 20% women! It’s amazing to have been a part of conference where community, generosity, and outreach were put front and center. I tried to do my small part by giving people directions during the tutorials.
Anyway, on to the specifics of what I did:
The first tutorial I went to was called “A beginner’s introduction to Pydata: how to build a minimal recommendation engine”. The intent of the tutorial was to introduce NumPy and pandas. I was hoping to learn some pandas-fu but I found the material poorly organized and didn’t feel like I was getting a good idea of why/when to use particular pandas features. The video for this one doesn’t seem to be up yet.
The second tutorial I went to was called “Bayesian statistics made simple” and this one was awesome! I was comfortable with Bayesian stats beforehand but a refresher never hurts and the instructor (Allen Downey) gave terrific explanations. He had a little Bayesian stats library for us to use in the programmatic examples, which was fun. (Though I had to re-compile NumPy and SciPy to get it to work. It used the one little corner of SciPy that’s often broken on Macs.) If you’re interested in learning more Downey is working on a new book called Think Bayes that you can read for free, Fernando Perez has posted his notebook from the course, and you can watch the video.
The PyCon Education Summit brought together educators from all kinds of backgrounds from K-12 teachers to those teaching adults. I went due to my interest as an instructor for Software Carpentry. Most of the discussion focused on teaching Python/computation in long-form courses to people who have zero programming experience.
I didn’t take much concrete away from the summit, but I was impressed with the sheer level of energy going into the Python/education nexus. There are many people out there experimenting with Python in education and developing lessons that use Python. There are also a lot of user groups around the country (like the Boston Python Workshop) that are actively working to bring new people into the Python world. Many people do this in their spare time! That’s the kind of community devotion I love about Python.
I gave a five minute lightning talk at the summit that was part a preview of my PyCon talk and part showing off ipythonblocks. The Notebooks for that are at nbviewer.ipython.org/5165758.
The first and most important thing you should know about the talks is that they were all recorded and the videos are online. There were about a million concurrent talk sessions and I’m still catching up on all the great stuff I missed. I highly recommend starting with the opening/closing statements from Jesse Noller and the Raspberry Pi keynote from Eben Upton:
I think there were standing ovations during each of those. And then there were the great regular talks I saw in person:
- Python: A Toy Language by Dave Beazley
- Do not miss a chance to see Dave Beazley talk. You will be thoroughly entertained and leave wondering why you do such boring things with your code. Here Beazley talks about using Python to control a hobby CNC mill.
- How the Internet works by Jessica McKellar
- Learn about the underlying structure and protocol of the web!
- Awesome Big Data Algorithms by Titus Brown
- Titus gives a great introduction to some algorithms and data structures that help deal with Data of Unusual Size. Also check out his blog post on the talk with links to his notebooks.
- Who’s there? – Home Automation with Arduino/Raspberry Pi by Rupa Dachere
- Rupa tells us how she built an automated front door camera. This talk was standing room only!
- What Teachers Really Need from Us by Selena Deckelmann
- Selena relates her experience getting to know teachers and how we as developers can best help them.
I gave a talk titled “Teaching with the IPython Notebook” that focused on how the IPython Notebook can help students learning Python. (Primarily by simplifying their interface to Python.) It seemed to go well and I’m really glad I did it! The video is up and my presentation notebook is at nbviewer.ipython.org/5165431.
I stopped by Simeon Franklin’s poster about making Python more beginner friendly and I was really impressed with the level of interest surrounding the topic. Even Guido was there seriously engaged in this discussion. With engagement of this magnitude at that level I think we’ll see people putting serious effort into making Python more user friendly right out of the box, which will be wonderful.
As Wes McKinney noted on Twitter, there were two things everywhere at PyCon this year: the IPython Notebook and Raspberry Pis. It seemed like every other talk and tutorial was using the Notebook and it’s no surprise, the Notebook is so fantastic for presenting code plus supporting material and then sharing it. It’s a major boon to Python.
Everyone at the conference (plus some kids who came for free tutorials) left with a Raspberry Pi. These amazing little computers enable all kinds of projects, often attached to an Arduino for talking to hardware. In Eben Upton’s keynote I learned that the “Pi” in “Raspberry Pi” is for Python since much of the system is built on Python. The site raspberry.io has been set up as a community of projects that use RPis but I’m sure Googling turns up a ton more. A small, cheap, low powered, easy to program computer just has so many possibilities! I haven’t had a chance to start hacking on mine yet but I’m looking forward to it!
A big thanks to STScI for sending me. Thanks to Greg Wilson for suggesting the talk idea and thanks to Titus Brown and Ethan White for looking over my proposal.