underscore_

Worth also considering all the extra times a kettle is boiled to make tea while waiting for the agents to complete

Thanks for the numbers, where do they come from btw?

Still thanks for sharing as it’s cool visualisation, especially if you already have dyslexia: completely unreadable!

Neat demo, but:

if we can increase our understanding and give the correct help to these people just think how many more Richard Bransons, Elon Musks or Boyan Slats we can produce

No thanks

To be fair I think fusion research is a great destination to siphon AI investment money to. I’m sure when Next Big Thing^TM^ comes along then they will dress fusion research in a Next Big Thing hat for further funding too

Edit: but also boo to data centre proliferation

And the report itself with an interactive map

Yup put it in the bin. Switching to Qobuz has been such a great change for me. Have been really enjoying finding new music again

A few from llvm (maybe?)

• llyfr
• llanc
• llif

There is still a distinction between something being public domain so not needing to ask permission and giving attribution which is always a nice thing to do.

79-95% cheating by source code look up, I sure hope these tools are not used in proprietary codebases attempting to reimplement anything GPL-like. Can’t wait to see the justification contortions of oh it’s fair use because my bot literally copy pasted the code, and the poor thing doesn’t know any better.

There are services like ORCID exist for uniquely attributing papers to an author. I think a federated review / publishing service could just provide a number of integrations for credential linking.

I was thinking about how such a tool might work over activity pub and I think it could work.

Thank you for sharing this, a great little TIL to go with my morning coffee!