3 ms·
I computed this list of top PDFs using the Internet Archive data. Basically it crawls the front page of HN and finds any .pdf link, records the number of upvot
by burtonator 7y ago
I computed this list of top PDFs using the Internet Archive data.
Basically it crawls the front page of HN and finds any .pdf link, records the number of upvotes, then sorts by score.
I did the same thing last year and it was a great way to find awesome/unique PDFs.
Ironically, the first post was about web scraping being legal and here I am computing the list via web scraping!
- inetknght 7y ago> it crawls the front page of HN and finds any .pdf link, records the number of upvotes, then sorts by score How do you deal with the numbers changing over time?
- burtonator 7y agoI take the max in a 24 hour period. I don't always have snapshots for all hours though so there might be some skew
- simlevesque 7y agois there a reason why you did not use the built in HN archives? https://news.ycombinator.com/front?day=2019-01-23 https://news.ycombinator.com/front?day=2019-01-23