Hacker News Reader

This Wikipedia article has a good summary of the replication attempts to date (including this paper).

Reading further below, I'm actually amazed by all the theoretical studies that have already been done from first principles (all apparently supporting the possibility of superconductivity). That's a pretty fast pace for science!

But have they asked chatgpt?

Magnus Carlson said that if you tell him there’s a winning move on a chess board he could find it very quickly, because his focus becomes extremely narrow.

I wonder if something similar is happening here. Since the scope was defined as LK-99, it becomes a narrow query instead of a broad one.

This is not just telling someone there's a winning move. It's telling them that X is a winning move and they just have to verify that X is indeed one. Many, many problems are such that it's difficult to find a solution but easy to verify one.

I know nothing about this but I stay curious. Just like Navier-Stokes equations can be proved with numerical approximations, can this be verified even if we never solve it?

True, but in this instance it’s my understanding that there were still some areas of guesswork where the paper wasn’t clear.

Reply on news.ycombinator.com