This.
Every time I think I’ve found a really clever, novel solution to a ultra specific graphics problem, someone else has already implemented it, usually much better than I’d have done, and was kind enough to put their code online for everyone to use.
DAMN ASSHOLES, being smart and kind, so I can’t even feel cool for 4 minutes, before I fail at implementation. >:c
Don’t worry. Future generations will have an opportunity to dive all these problems after AI destroys a whole generation of useful development along with all the repositories of prior research.
There will be a whole new class of historian and archivist dedicated to sniffing out slop.
Well they also probably spent well more than 4 minutes trying to implement it and failed multiple times along the way. Just make a small tweak and call it your own, let other people get that pesky trial and error phase out of the way.
This “Et Al” fella has stolen all the best ideas!
I feel bad for kids named Et Al. No matter what field they go in, they will always be compared to the original Et Al.
Yeah but it’s not like Et Al. has been supervising/leading any papers, research or stuff like that. Their name is always stapled on the end.
But have they tried to solve it with AI?
Them: Look at me I’m the AI now.
Unironically a lot of AI and LLM research was done in the 70s
We’ve had several implementations even that were most of the way there just limited largely by hardware constraints. Megahal from the late 90s is pretty similar to modern LLMs, just drastically simplified.
Let me guess, it’s Schmidhuber
This was me once upon a time but with making stuff on YouTube. It lost it’s glimmer and I wasn’t even really trying to make money from any of it. I could still put out some stuff but It’s not a priority for me anymore.
