"Tran argued that he responded reasonably to a perceived sudden emergency caused by the passengers' prior excessive drinking, of which he said he was unaware, & that he could not anticipate that Normandin-Parker would leave his vehicle and wander into traffic."

You think they don't?

Reduce means to take away.

But reduce isn't taking something away. It's combining multiple things into one -- squeezing multiple things together. Reduce is poor verb for that.

It should've been called squash. Just like squash-merge.

You need to think in terms of supply and demand.

The demand is there, but the supply is also going to skyrocket. Free open weights models will contribute to supply too.

There will be a new equilibrium that’s hard to predict.

- it's first image is way earlier: 2KB vs 8KB

- from 8KB to 100KB it's blurrier than AVIF, but it's more pleasing to look at and less blocky than AVIF. I highly prefer it.

- from 100KB until finished 288KB it's actually ~4x sharper than AVIF

Overall JpegXL gives the much nicer experience.

^ tested with pokebowl (https://jpegxl.info/resources/progressive-loading-demo.html)

This is what we should’ve been doing a long time ago.

- I prompt better when I can understand the code, when necessary

- I can catch wrong assumptions it made more easily

- Good code helps the AI too, AFAICT

Brent Underwood has spent $30K+ trying to get water on his land: https://www.youtube.com/watch?v=atKOoMvLmTA&t=1s

If we are going to have a government promised right to water, we can't just ignore people that move to places where it's hard to find water.

A lot of rural cities don't even provide water. People have wells and you pay for them yourself.

Does it complete the sentence?

If not then it’s not a next token predictor. Or at least not a good one.

For use as training data you still need some right to it AFAICT.

No good way to get around that except buying an Apple device and taking its key.