• 3 Posts
  • 74 Comments
Joined 1 year ago
cake
Cake day: June 30th, 2023

help-circle

  • IMHO 2012 was the one that really broke their brains. The tea party types tried to get various firebrands on the ticket, but end up having to support Romney. Hey, at least he’s a squeaky-clean telegenic millionaire pushing the most severe conservative fiscal policies! Proceeds to get stomped by the Obama campaign so bad that Karl Rove couldn’t believe it as it was happening. THEN the establishment GOP flirts with moving to the center on immigration. The backlash against that on the right, supercharged by the mainstreaming of mobile social media (plus social justice protests and the looming Clinton campaign) was what fueled the rise of Trump.



  • It’s just not practical – no Minecraft server or map can realistically hold all the books in the Archive, or even just the 500k that were removed. Even if it could, you’d only be able to read them by literally taking your avatar to the book object and reading it in the tiny in-game interface.

    The Minecraft thing is just a gimmick to promote awareness of press freedom and censorship, not a plausible way to deliver books to people. If the IA wanted to “set books free” they’d be better off using torrents or something like Libgen (and even then they’d still be criminally liable for making the files available, even if the publishers couldn’t stop the files from being shared further).







  • If it is just a repackaging of ChatGPT’s existing “search the web” function, I don’t know why they’d bother. It can at best summarize a page of search results for a very literal-minded query, and even then it’s often lobotomized by the fact that OpenAI has made it easy for a large number of top websites to opt out of having their pages accessible to their search crawler, which means you’re only getting a summary of the search result snippet and metadata. A competent user of Google search can run rings around it in terms of research, even with Google’s decline in quality. I guess it makes it faster to answer basic queries for recent information not in the training data, but that hardly seems worthy of a big event.