Plotting eight years of book buying habits

Since late-2008, I’ve been adding my book purchases to my online catalog at LibraryThing. I thought 8 years worth of (a relatively small set of) data would be worthwhile to play with.
They had an export option that gave me a tab-separated dump of all these books. One column in this file is the ‘Entry Date‘ that shows when I added it to the website (usually within a day or two of buying it). This is what I was interested in.

My first exploration involved matplotlib. It did its job well but I got sidetracked into ‘prettier’ packages like and bokeh. The latter is what I ended up using.

The data processing was trivial. I only needed to calculate how many books I added in a given month and plot a bar chart from the resulting counts. The result looks like this:

And it confirms what I suspected!

  • I got married in 2011. The density drops off drastically then, but is still reasonable.
  • Kid #1 popped out in April 2014, and the second little fellow in March 2016. The counts are a lot more sparse there onwards 🙁

I’m trying to get them interested in books so they’ll leave me in peace as well. Let’s see how that goes.

…and here is the code to make it happen. It’s just a simple script so I didn’t make the extra effort in packaging it and so on. I used anaconda/Spyder to develop the script, and it was pretty easy, despite my complete lack of knowledge in this area.

Leave a Reply