"The new bound saves a factor of ≈ log_3(n) over the prior record by Ford-Green-Konyagin-Maynard-Tao from 2018. The result is also now formalized by Alexeev in Lean."
I've been told that prediction markets provide reasonably good forecasts compared to most other methods. Could you help me understand why that might not be true?
This is one example from a source I just happened to read this morning, which is otherwise pretty positive about the prediction markets.
> Beyond fears about bad actors attempting to distort the markets for profit, some political operatives say the sites rarely deliver political forecasts that are different from traditional polling.
> For instance, polls during Wisconsin's gubernatorial primary showed Democrat David Crowley lagging other candidates by wide margins. Kalshi similarly projected Crowley as a long-shot candidate — but Crowley landed a surprise win.
> "It appears that there really aren't any predictions here, they're just reacting belatedly to public information, like polls and news stories," said Eddie Vale, a Democratic strategist in Washington who worked on Crowley's campaign. "If it was just a fun entertainment tool for political junkies, fine, but to me, all of this is a marketing sheen for an online gambling operation."
So it's not really that they provide bad predictions, it's more that they don't offer anything new. Except, of course, in the case when someone is using insider information in an attempt to profit. All this while also having the downsides of skimming money off every transaction and encouraging gambling habits.
This. Generating a frontend with guarantees can almost not be done with any other language. If you use the graphql module with code generation it's an absolute game changer. Every time I have Claude generate a moderately complex frontend in react I regret it sooner or later due to state issues and inconsistent behaviour. Not with elm. Everything that makes it difficult makes it perfect for ai.
> For example, is it possible that unknown to the public there were multiple launches of artificial satellites long before Sputnik with some launches timed to coincide with U.S. nuclear tests? Or rather, do the current findings represent detection of a non-human technosignature? Due to data limitations, such hypotheses cannot be subjected to falsification.
This is why we need better education in this country.
Anyone who has read any amount of history from this time should know that this is very simply not possible.
This statment is unfalsifiable in the same way that "I'm a six-legged alien from Venus typing this message from orbit" is unfalsifiable. It's just flat nonsense.
This actually _is_ falsifiable; I've written before that you can determine whether the point-spread function of the transients match the PSF of stars. If not, then it does not come from outer space, it's an issue with the film, i.e., radiation.
reply