Building A Custom Reading List With Airtable

February 17, 2020

I built a custom book tracker and reading history page, and now I’m kind of in love with Airtable.

Airtable

Airtable is an online spreadsheet app with a relational database’s heart and wonderful first-class API support. You can build a huge array of useful stuff using their tools: their product page and example universe are a good showcase of what Airtable can do. They’ve got an incredibly generous free plan, too: everything detailed here fits in a free account!

For hobby-stage products, I think Airtable is a lift in productivity on par with hosting platforms like Heroku. It’s a huge force multiplier for prototyping or building home-cooked apps tightly fitted to your own workflows. The first stage of this reading list was up and running on its own in a couple minutes after just setting up the fields in my Airtable base: a spreadsheet-like interface to my data that was powerful and visually pleasing. More importantly, it was already useful and useable. I used this first stage of the project as-is for a week, to track my read books and note ones I’d like to read in the future. I got a sense of what working with the modeling was like, and could remove and alter fields with no ceremony. Before I imported my full reading history or wrote a single line of code, I knew the structure was well set up for my personal use.

This is all heavily inspired by Simon Hørup Eskildsen’s work, both Minimum Viable Airtable and his excellent airrecord Ruby gem. Big thanks to him for the spark that made me start all this.

Book Tracking

Why build my own reading tracker, after a decade or so on Goodreads? I wanted:

  • An easier way to filter my to-read list by author background. Specifically, reading a lot of authors with diverse backgrounds is important to me, so I wanted my to read list to be able to show me a single page of all the books that aren’t by straight white dudes.
  • The flexibility to show reading history and stats on my personal site exactly how I wanted them. I do a summary of author backgrounds among the year’s books in my yearly review posts. This was a somewhat manual process, and I wanted to have the breakdowns automatically tracked and summarized for current and past years.

I only wrote code here for the second part: the reading history page. Just building a filtered Airtable view was good enough for the to-read list by author background, and I didn’t need to go any further than that!

For the reading history page, a little context: this site is written in Jekyll and hosted on Github Pages. There’s a limited set of plugins you can use with Pages’ basic Jekyll integration, so I use two repos: a source repo with the Jekyll code and markdown for blog posts, and a built files repo with the compiled HTML to be served. I use a simple publish script to sync content between them.

The reading page is built with a relatively small amount of code. I have a Ruby script to query my Airtable’s API for my reading history, to which it does some light processing and dumps the results out to a single JSON file. The template for the reading page simply loads that file and renders the page with some straightforward Haml. This process is run on my laptop right now—maybe someday I’ll automate it, but at a couple books a month, remembering to run a single shell script after updating my Airtable reading records isn’t exactly a huge burden.