Thanks to enhanced pitch tracking in the last few years, they can now display even more information. The Pittsburgh Pirates' first- and third-base ribbon boards show pitch speed, vertical and horizontal break, and IIRC even the name of the pitch (bucketed based on the speed and break characteristics). It's a really neat addition to have in real time.

Imagine how much less you would use text messages if they still had a per-message cost.

The obvious problem is what happens when operations become abnormal. ATC shouldn't be staffed for normal operations, because then abnormal operations lead to catastrophe. Welcome to last night: the weather is bad, which causes a plane to abort two takeoffs, which causes that plane to need emergency services. This increases the controller's workload beyond his capacity, so he accidentally clears the emergency vehicle to cross in front of a landing airplane, and they can't see the airplane because the weather is bad, so they follow the instruction and promptly get hit with an airplane.

When some bad weather can be the difference between "this is fine, one controller can handle it" and two dead pilots, you need to be staffed for bad weather.

It only works if you deploy it against someone lower-status than you. The tactic is largely irrelevant and can be seamlessly replaced with any of a number of other tactics as needed. It's just enforcement of power hierarchies.

Polymarket and Kalshi allow you to monetize almost any outcome (as they themselves will tell you is the goal). Therefore, there is profit available if you can predict the outcome better than other people, or if you can influence the outcome. But I don't think people have really noticed that profit is also available if you can influence other people's predictions to manipulate the price and sell at a profit. Spread disinformation. Suppress information that hurts your position. Sending death threats to journalists? Sounds like an avenue to accomplish that. Want to do that at scale? Agentic AI can help.

We need to severely restrict prediction markets, and soon, or there's going to be a lot of adversarial activity at scale, and we can't always predict what kind of activity that will be.

Too often, advocates try to smuggle in their preferred policy using stories like this as cover.

Procedural generation is good when variety matters more than quality, which is a relatively rare occurrence.

I've found that this is also true of American engineers, particularly those fresh out of college. So many people have internalized that open curiosity will yield no result at best and direct punishment at worst.