tomthe

I thought you were referring to this slightly humorous video about the same topic: https://www.youtube.com/watch?v=7dO9Lm_CXz0

Cool! I made something similar with the same datasets a year ago, but with geocoordinates instead of timestamps: https://theilemail.de/wikipage/ It shows a zoomable map with the "most important" wikipedia entries at specific locations.

It was very hard to get the parameters right so that the text is always displayed in a good size, while showing not too few or too many entries. I never published it until now, I think.

The creator of UMAP and PyNNDescent tried to let Opus 4.5 translate PyNNDescent from Python to Rust

No, it is not as good as Veo, but better than Grok, I would say. Definitely better than what was available 2 years ago. And it is only a 7B research model!

Yes, of course I do ignore a lot in that calculation. I just wanted to calculate the biggest possible usefulness of this cable. Especially the resistance losses could be quite disastrous.

To some degree, yes. I just looked it up and the EU produced ~2500GWh in 2023, which is around 280 GW on average.

If I (and only I) owned such a cable from Europe to the US, how much money could I make by buying cheap solar energy from the bright side and selling it to the dark side of the Atlantic?

First thought: 10 GW * $0.03/kWh 4 hours/day = $1.2Mio per day [0]

I am not sure about my assumptions...

[0] https://www.wolframalpha.com/input?i=10+GW+*+%24+0.03%2FkWh+...

I wonder if you could implement it with only static hosting?

We would need to split the index into a lot of smaller files that can be practically downloaded by browsers, maybe 20 MB each. The user types in a search query, the browser hashes the query and downloads the corresponding index file which contains only results for that hashed query. Then the browser sifts quickly through that file and gives you the result.

Hosting this would be cheap, but the main barriers remain..

Very cool open release. Impressive that a 27b model can be as good as the much bigger state of the art models (according to their table of Chatbot Arena, tied with O1-preview and above Sonnet 3.7).

But the example image shows that this model still makes dumb errors or has a poor common sense although it read every information correctly.

It is not sufficient to understand the content very well, you also have to understand the state of the mind of your pupils very well.

I think there are pretty strong traces of pro-china manipulation on Tiktok.

See https://www.nytimes.com/2024/04/24/briefing/tiktok-ban-bill-...

Content about something like Hong-Kong protests or the Tiananmen square are suppressed.

Can you elaborate? I had no problem with DuckDB and a few TB of data. But it depends on what exactly you do with it, of course.

Wait, you eat only beef? Or beef as the only source of meat?

anymore??

I made a similar map but with tiles that only load of you zoom in far enough: tomthe.github.io/hackmap/ (Sorry for posting my link so often) That way it has to load only a few megabyte for the first view.

Nice introduction, but I think that ranking the models purely by their input token limits is not a useful exercise. Looking at the MTEB leaderboard is better (although a lot of the models are probably overfitting to their test set).

This is a good time to chill for my visualization of 5 Millionembeddings of HN posts, users and comments: https://tomthe.github.io/hackmap/

Thank you, I really like the default tutorial how one can play with it. Is it possible to visualize data with this?

HN user ggerganov made a similar tool with real historical numbers from Bulgaria:

https://lottery-check.ggerganov.com/

This one might be interesting for you: https://daylightcomputer.com/ it has a fast e-ink-type of display. Unfortunately it is expensive and not available.

Yes, that seems to be the explanation. I found a few pictures here: https://www.komoot.com/de-de/highlight/615007 but they also do not show how it works.

Maybe 750 years and 15 min.

-=[ iOSnoops - 2010-2023 ]=- We would like to inform you that we have decided to shut down the site.

The reason? It's simple, the economics are not there anymore, we can no longer run this site without massive out of pocket expenses.

After long deliberation, we decided it's best to simply let go.

A big thank you to all our fans and viewers!

And good luck to all the copycat sites that shamelessly scraped and stole our content on a daily basis... no harm, no foul, it's how the web works!

It was a fun ride, but it's time to say goodbye.

Until next time!

I thought this simple and short statement was remarkable enough for a submission here.

Sorry, you are right and I stand corrected. I must have misremembered this. Here is a tweet from PG correcting this: https://x.com/paulg/status/1796107666265108940 via https://news.ycombinator.com/item?id=40521657

I originally got this sentiment partly from this: https://news.ycombinator.com/item?id=38378216

Why do you think they might be wrong when they want a simple website?

I agree with the no, but you can make space itself rotate because things in space rotate: https://en.wikipedia.org/wiki/Frame-dragging And that in turn would rotate things in space... or not?

Not all (look up how he got fired from ycombinator). But as long as there are enough new people who want to play with him, he might do fine (and there are a lot of willing VCs). Once the word starts to spread about him, through stories like this, he might experience some backslash.

I would love to play some board games with Sam Altman. He is smart, strategic and very persuasive (I heard). He would excel especially at games like Risk or Mafia, where you have to persuade your opponents to collaborate with you.

But after a few rounds, no one would help him anymore and he would loose all the games.

I made something very similar a few weeks ago. I also included usernames with the average of their comments: https://tomthe.github.io/hackmap/

Wait, how is this possible?? I tried to visualize a few tens of MB of text in the browser [1] on a canvas and it gets really choppy if I don't optimize it by not showing stuff that can't be read. Impressive that it can handle gigabytes of text!

[1] https://tomthe.github.io/hackmap/

And the pentagon orders randomized amounts of pizzas every few random days.