> It brought me joy to help people figure out why their #ASR code wasn't working, or assist with a #CUDA bug.<p>I don't understand why they would want to remove their code instead of wanting it to live on and further help more people through AI. Just because OpenAI charges $20 a month? It takes a lot of electricity to produce results from AI. Just seems shortsighted on flimsy principle. If it were me I'd wait for the lawyers to sort out what SO and OpenAI are doing with CC-By-SA instead of making a big stand about it on the internet, but I can sort of get it I guess.<p>Edit: I called it flimsy principle because it's not clear that CC-By-SA prevents this until the court rules
Related:<p><i>StackOverflow is banning accounts that delete answers in protest against OpenAI</i><p><a href="https://news.ycombinator.com/item?id=40297027">https://news.ycombinator.com/item?id=40297027</a><p><i>Stack Overflow users deleting answers after OpenAI partnership</i><p><a href="https://news.ycombinator.com/item?id=40302792">https://news.ycombinator.com/item?id=40302792</a>
I don't know when it was most recently changed, but...<p>"and you grant Stack Overflow the perpetual and irrevocable right and license to access, use, process, copy, distribute, export, display and to commercially exploit such Subscriber Content, even if such Subscriber Content has been contributed and subsequently removed by you"<p><a href="https://stackoverflow.com/legal/terms-of-service/public" rel="nofollow">https://stackoverflow.com/legal/terms-of-service/public</a><p>This content isn't licensed under CC-BY-SA <i>to stackoverflow</i>, it carries additional riders which grant SO the legal right to sell it however they please.
This whole ordeal reminds me that I actually do need to go back and clean up and edit all my stack exchange posts.<p>Not because I want to protest but because they are bad. Even highly upvoted and accepted answers of mine are pretty poor quality, don’t give the best advice, and in some cases provide misinformation. I am very sure I instructed people on how to make Frankendebian in more than one answer. And I can’t be unique in this regard.<p>Just scraping Stack Exchange without someone reviewing what that information actually is, is going to actively feed bad information into the model.