Like, there are entire journals and programs of study devoted to it. I’m not just guessing here.

That's me, and Google correctly on the first page finds ONLY pages related to me.

Kagi pollutes the page with many results for my doppelgänger, the much more famous Daniel J Drucker of the University of Toronto, even though none of those results mention mclean.

That is literally the thing Kagi was supposed to be better at - actually respecting your search terms instead of thinking it knows better!

Using the Census ACS age brackets, about 20-ish% of louisiana's population is under 15, and another 20 is between 15 and 29. Everyone 30 and older adds up to the other 60.

So a hair over 60% are were at least 7.

But that's who lives there now not who lived there then. Between 2005 and 2006 the state population dropped by 6% and most of that displaced population never returned - people coming in from elsewhere weren't there for Katrina. So the fraction who were both living there AND old enough to remember it is considerably smaller than 60%.

So like I said, roughly half.

The linked article is from research done more than 4 years ago. If you're basing your idea of what LLMs can or can't do on what they could or couldn't do in 2022, well, good luck to you.

I think the reason is my searches are almost entirely long-tail searches that Kagi's index just isn't good enough for. I am never searching for something like "best mattress" or anything else that is heavily SEO'd - it's always something very specific - so the result page in Google looks pretty much exactly like the Kagi page, only it nearly always has the result I'm looking for where Kagi's doesn't.

I'm not entirely clear on the mechanism by which memories make it into context, so it's possible some of it isn't all the time, but it does seem to be working reasonably.

Again, it's not as good as Claude when it comes to writing "not like an AI". But it's significantly better than it was.

I also put together this ridiculous thing[1] because I missed the font and color scheme of Claude.

[0] https://gist.githubusercontent.com/dmd/91e9ca98b2c252a185e8e...

[1] https://github.com/dmd/aimpostor

Wish you could get that soothing Claude font and color scheme?

Or maybe you just hate native apps and prefer the bloat of Electron?

Well it's your lucky day, because this is an Electron ChatGPT wrapper that scratches your itch. It even has dark mode. (Applaud here.)

Comments URL: https://news.ycombinator.com/item?id=48135828

Points: 7

# Comments: 2

In fact I would generally bet thousands of dollars against that.

How many crowdfunded projects do you know that have raised even one percent of that? Who’s going to be in charge of collecting that scale of money? Perhaps some sort of company formed for the benefit of humanity, which will promise to be a non-profit? Some sort of “Open” AI?

Oh, wait.