2019-01-28 UTC
# 
[Vincent] I stored them separately (rather than in one file) because there were complicated race conditions if a site sent you multiple mentions at once. You would be updating the file to add the first mention while the second/third etc would try to access the file and fail. I didn’t want to build a complicated queue system to manage this, so I store each mention based on their ID and use the webemention datestamp as the file location.