Notebook
Here's where we post periodic updates on what we've been up to at Fathom. Reflections on the interesting stories that emerge from our client work, side projects, after-hours rabbitholes, and other miscellaneous threads of inquiry.
You can also follow these posts as a feed in your feed reader.
Here at Fathom, education and access to technology are huge priorities. This informs many of our efforts, from the classes we teach, to hosting guests visits from groups like Girls Who Code, to our work with the Processing Foundation. Bringing clarity to complex ideas and difficult explanations is core to what we do.
It’s been a busy couple of months, and we finally found the time to write about it.
Over the last year, we’ve been working on ways to understand very large document collections—hundreds of thousands or even millions of documents. We began using Wikipedia as a source collection because it has millions of documents, but it’s also an openly available resource representing a vast range of recorded knowledge. It’s huge, it’s free, and we’ve always been curious to learn the breakdown of what’s being documented. What receives the most attention?
Anisha’s analysis of video archives, FBI documents, and Wikipedia
Visualizing 200 years of growth in finance, dissecting Russian ads in “Fakebook,” and disassembling image data
As part of our exploration of ways to understand large document sets, we took a look at the Facebook ads funded by the Russian government between 2015 and 2017. They aimed to stoke tensions by targeting polarizing messages at both sides of controversial issues.
“Paying the President,” visualizing oral histories, and kicking off a summer of cell biology
As a studio with its fair share of computer nerds, we greatly enjoyed digging into the interview archive of the Computer History Museum. Using code, we extracted dates and entities from the 800+ oral histories to create a personal timeline for each interview.