3 ms·
Someone recently pointed out the Aaron Schwartz was threatened with going to prison for scraping, meanwhile there's hundred of billion of dollars right now inve
by nailer 2y ago
Someone recently pointed out the Aaron Schwartz was threatened with going to prison for scraping, meanwhile there's hundred of billion of dollars right now invested in AI LLMs build from... scraping.
- ToucanLoucan 2y agoRules for thee and not for me. Same as it ever was.
- startupsfail 2y agoAnd Markdown is the primary format.
- bdjsiqoocwk 2y agoWhy go to AI LLMs? Scraping and indexing is all Google search does.
- stavros 2y agoThat's because the megacorps can scrape you, but you can't scrape the megacorps.
- jahewson 2y agoJSTOR is a non-profit run by academics. Indeed, they have plenty of money, but they’re no megacorp.
- SJC_Hacker 2y agoGenerally it should be "More powerful entities can scrape you, but you can't scrape them back" Google scraping JSTOR (hey, don't they do that already with Google Scholar?" is much less of a problem then JSTOR attempting to scrape Google.
- TiredOfLife 2y agoWasn't he scrapping paywalled stuff?
- nojvek 2y agoAaron didn’t have an army of lawyers that Megacorps do. He took the papers and put it public. Blatant copy right violation. LLMs are in gray waters of derative work, not verbatim copy of original. Different judges have had varying rulings.
- sanbor 2y agoMy understanding is that Aaron just downloaded the data, not that he made it available to the public. That’s what I found also in Wikipedia[1]. [1] https://en.m.wikipedia.org/wiki/United_States_v._Swartz#endnote_quantity_downloaded https://en.m.wikipedia.org/wiki/United_States_v._Swartz#endn...