Always learning

What I learned form speaking at PyTN

File under: pytn python

PyTn was my first technology conference and it was incredibly fun! I met some great people, I sat in on some incredible talks and picked up some sweet swag. And I definitely have the speaking bug. But the best thing about the experience is what I learned.

I learned to be more self-confident about the knowledge and experience I have. I’m still starting the journey into development and data science, but I have a bit more experience than some people and that always needs to be my target audience: people that may not know much about my topic. The days of being a computer guru are over, the field is too wide for anyone to know it all and so there will always be someone who might not have had the chance to pick up what you’ve already learned. Not every talk needs to be deep and filled with arcane wisdom from decades of research. A bit of depth and a bit of breadth in a well written talk and the audience will come away with something useful. So, don’t worry so much about presenting something that people already know; if they know it, they’ll probably go to a different talk and the people in your talk will know what they came for. Deliver on that promise and things will go well.

I learned that I have to stop working the same way I did in college, somewhat. In college an all-nighter to write a paper was a common theme. I’d do a lot of research and take a lot of notes, I’d form arguments in my head (I was a philosophy major, that’s how we roll) and lay out my story end-to-end, but I wouldn’t write a thing until the day (usually night) before. This talk was a hybrid of that process: lots of research, lots of notes, lots of talking my story out in my head. But I also wrote some “drafts” along the way. Since I did this in iPython notebook, I was able to easily keep notes and code all together and change them as I needed, which helped keep me focused. But, I learned I need to do better about committing work along the way. Thinking it all through and trying to put together a full story is how I work, but I need to get that narrative on paper much sooner as a professional. I did better than I would have as a student, but I still put more stress on myself than I would have by being a little less up in my head and a little more in the working code.

Probably the best thing I learned, though, was the topic I presented on. I did this work a few years ago as a grad school project, but I did it in Java then and kind of hated the process; munging data in Java just shouldn’t have to happen. The learning process for pandas, especially in an iPython notebook is pretty fabulous because you get the immediate feedback that something like TDD proposes, without having to build up an entire infrastructure to pull in your data and twist it around to get what you need. I don’t think I’m exaggerating when I say a project that took me half a semester in Java would have taken two weeks using the python tools I talked about. And I didn’t even try to visualize the data in Java; I’m sure I could have, but the way these tools interconnect made it really easy to get a few graphs up to show me where I stood with the data. So, what I learned wasn’t as much about the concepts (that was more about refreshing memories that were a few years old) as it was about the great tool set.

Lastly, I learned that I need to keep doing this kind of thing. I don’t just want to do it, I need to do it. Helping people learn is one of the things that fuels me, makes me feel like I’m giving back in a positive way, and helps me focus on what I need to learn next. This blog, my twitter, talk proposals; I hope to use all of these to be more engaged in technology, development, and data science and most importantly, to give back to and learn from these amazing communities.