2019-10-23 UTC
# 
[tonz] For a Twitter harvesting script that checks which things are linked to with a given #topic I check those links for their nature, as far as possible. I gave up figuring out if I was looking at a post or a page there, only now check if it’s a domain name only, or something with more behind the tld/ and if there’s a specific reason to assume it’s a vid, preso, pdf etc.