An Interactive Atlas of Objectivism
Incomplete but Interactive Network Graphs and Trivia Quizzes
I’ve drafted and deployed several interactive entity-relationship (or “network”) graphs of the Objectivist movement. They’re all incomplete and under active development, but you’re welcome to try them out and message me with any corrections or additions that you’d like to see.
Context
A year ago next week, I wrote one of my first Substack articles in an attempt to find other Substack blogs that had been publishing on Ayn Rand or her philosophy of moral objectivism. The motivation was to help me learn the “landscape” of a topic of interest within the platform, using techniques from my career as a data scientist. To that end, I wrote an RSS-based data search and scraper to help me find them.
The results were incomplete and noisy, but it was a mildly useful first attempt. I published them within the article as a list of titles, authors, and hyperlinks.
I ended the article with a comment or footnote indicating my intention to follow up with a more complete, cleaner second pass at data collection. Thankfully, I didn’t mention a timeline, so my credibility hopefully hasn’t been called into question.
Although that never materialized, I started building a database of people, entities, and facts related to objectivism, for the purpose of self-education, and the scope of that project has grown large enough to bring us full circle to updating data on Substack as well.
The Database
The database consists of information on objectivists, their organizations, related Substacks, and trivia. I began with customized data scraping of the websites of the Ayn Rand Institute, the Objectivist Standard Institute, and The Atlas Society, in addition to blogs from my “Objectivism on Substack” article.
The programmatically collected data was both very messy (lots of mistakes and inconsistently typed proper names), but I have been manually cleaning and reviewing it. The process is ongoing, and some organizations are still missing records showing connected individuals. Corrections and contributions are welcome.
The trivia component of the website followed a similar pattern. I initially wrote a script to batch process large chunks of raw text from all of Rand’s works, which were used to generate multiple-choice questions and store them in the database. It generated 8,000 questions.
Unsatisfied with the quality of too high a proportion of those questions, I made a decision to limit the quiz to only manually reviewed questions. I’ve only reviewed 200 of the questions at the moment, which are available in the quizzes on WhoNeeds.it, but hopefully this number will increase over time, and I am considering opening up the question review tool to other users for that reason.
What’s Available
At the time of writing, you can access my Python-based network graphs of people, orgs, and substacks on WhoNeeds.it. I built that site using GitHub, which has locked me out for nearly a week. For that reason, I’m unable to edit the code behind the graphs and instead decided to create a separate enhanced version using the R language here.
For those with an interest in AI code generation, note that the Python version was generated using GitHub Co-pilotand the human (me) prompting, testing, and correcting errors. The R version is completely human-written, on the other hand. Surprisingly, they both took about the same amount of time1. Let me know if you find quality differences.
The newer graph supports scrolling, zooming, filtering, images, and external links. It does not (yet) have a separate tag for Substack.
To be fair, I had a much clearer idea of what I was trying to accomplish when I wrote the R version, which probably saved me time.