2 comments

[ 3.0 ms ] story [ 10.5 ms ] thread
Very intriguing data set. Would love to see more.
This is a beautiful use of revision control; the dataset is intriguing in its own right, but changes to it are extremely insightful. A+

This also exposes subjective issues with diff algorithms, like Github's default. Viewing this compare [1] for the file 'statement.txt', I would strongly prefer smaller runs of text to be detected as identical, rather than it showing a large block as being different. Of course, I realize this is entirely outside of your control.

[1] https://github.com/fomc/statements/commit/65982075a5613fcfa4...