dubslow

Yes, that's what I meant. Our current zugzwang detection code has no statistically-significant elo cost compared to *not* doing it, as shown by that test. (If the zugzwang detection *did*...

Toxic has a bunch of [sounds](https://github.com/Tox/toxic/tree/master/sounds), but of the two I listened to, I didn't like IncomingCall and much prefer qTox's notification to RecvMessage.

Nope.

> The current code is not very robust if for some reason `complexity` is negative in some scenarios, as this gives a negative `totalTime` ? > It was changed from...

Perhaps this is dumb, but why do we call `collection.count_documents(q)` when we've already called (or will anyways call) `list(collection.find(q))` ? Especially when the list is filtered further after the `find(q)`...

A Discord user claims that weird shuffling can be found even in simple cases of KRvK, and that Stockfish 14 does not have a shuffling problem, while SFs 12, 13...

> In many cases you want partial results from subtasks as soon as they're available. (Possible solution: Use a memory channel.) Of course I'm only talking about retrieving return values...

Disclaimer: am noob, but I don't see how that makes the result available outside the `with nursery:` statement. Indeed as I edited into my previous comment, in principle our code...

Here's a toy example expanded from a throwaway in the docs. Obviously it doesn't run, and as I said in my previous comment I think it essentially doesn't work as...

And modifying that, I've come up with a toy example that actually works and does what I want it to: https://gist.github.com/dubslow/379342c6dbc0efc69491fe9c8becc089 However I find this code to be extremely smelly,...