Everything is a tradeoff.

The hardest issue I’m working through right now is the tradeoff between “get me what I expect to see” and “delight me with something I wouldn’t have otherwise found.”

Unfortunately, solving this is less “use rigorous math to find the optimal solution” and more “taste the soup as it’s cooking and adjust on the fly.”

I put out a bunch of changes today which I think will get me closer to the right ballpark. I’m going to watch it for a few days and see how it plays out. The good thing about U-curve optimizations is that you have a lot of space in the middle to get it approximately right.

Please let me know if you notice things improving or getting worse.

Thanks,

Jason