I like to write. It’s a hobby and a passion of mine. When I was a kid, I lost myself in my own fantasy novels. I published a supernatural suspense anthology under a pen name in my early twenties. I still, from time to time, add bits and pieces to a heist novel that has grown in my head while I ride the Metro to and from work.
So there’s the writing part. But why the heck have I decided to write about data science?
Learning to write is learning to think. You don’t know anything clearly until you state it in writing.S. I. Hiyakawa
I’ll admit, I haven’t been at the data science game all that long. I started out as an operations research analyst and a complexity scientist. However, I firmly believe there isn’t one field out there you can find that won’t benefit from understanding data science. I figured out fairly quickly that it was a skill I needed to include in my toolkit, and that I needed to learn quickly.
As I learned, I wrote a number of different pieces – information papers, conference presentations, the occasional journal article, publicly shared blog posts on Medium and LinkedIn – on learning data science. I figured that some of the things I learned along the way might help others needing to learn and provide a useful guide.
But the more I write, the more I’ve learned. And the more I publish out in public forums, the more feedback I’ve received and the more collaboration opportunities I’ve discovered.
So here, simply, is why I write about data science.
I write for understanding.
My thesis adviser often told me that if I fully understood my thesis, I could explain it clearly to every drunk at the bar on a Saturday night. So as I struggled to write about complicated concepts (I wrote my thesis on creating a process to integrate probabilistic design and rare event simulation to solve problems in high reliability systems), I began finding new and interesting ways to explain them to others.
That taught me that the more I force myself to write about something, the better understanding I gain. It’s easy to copy and paste code, and while you know what it does, you don’t always have a thorough understanding of how it works. Writing about it, whether it’s a critique or a tutorial or some other explanation, forces you to really learn what it’s about.
I write for the feedback.
It can be ridiculously hard to get honest feedback on your work, especially when you’re largely self-taught. You miss out on having a professor grade your homework. So while you have developed a largely effective algorithm, you don’t have anyone around to tell you that you used a far more complicated method than you should have, or you made an error in your code that confounded your results.
There are lots of folks out there digging into the field of data science with far more experience than I have, and they have been great about providing insight on how they learned, what works and what doesn’t, and where the state of the art in our field really lies.
I write to educate my bosses and myself.
One of my pet projects is the education of senior decision makers in my industry. They’re savvy folks. They know they need to employ data science and data analysis techniques, advanced analytics, and even machine learning and AI – but they’re not really sure what those are and not really sure how to best use them.
Writing about data science helps me put complex concepts not only into simple terms, but to put them into the context of the larger organization and how the knowledge of data science can lead us to new horizons.
I write because I believe it’s a necessary skill.
I believe great techniques, ideas, and concepts are only as good as your ability to communicate them to others. There are a lot of great data scientists and analysts out there, but while they can make code stand up and dance for them, they don’t really care all that much about explaining how it works to others.
Unless you’re King of the World for the day, you need to be able to explain how your analytical techniques work and defend your results to a wide variety of audiences who should be questioning them. Writing about data science has helped me practice this on a routine basis.
I find that writing about various technical subjects increases my understanding, helps me share knowledge, enables me to get feedback I might not normally get, and makes it a whole lot easier to produce information papers and reports when my job requires it!
So that’s why I’m here. Your turn to share! What brought you here to read my thoughts on data science and other subjects?