[yogo]

I can sort of see this for physics but what if it's programming or software in general. I wouldn't be able to tell if someone is good if they were describing something to me in F#, as an example. They could be going about it the wrong way.

What if someone was really onto something and they were describing it to you. They might be trying to explain it as simply as they possibly can and you might think you understand, but a lot could be lost and your impression might be that, that person is not good at what they are doing or that it's a toy.

It all comes down to good communication but I think overall, we learn more over more time. I get that there is a time constraint with the current approach so maybe to scale things there should be some sort of engine that processes people/teams from a watch list over a period of time. For example, mine data from a private YC journal that applicants can start writing in a year or six months in advance. I don't think that should be a problem since a lot is submitted in the applications (my basic understanding of the process).