spullara

They included someone who died on the way to Reno hospital this year.

Glad you are on the case before these companies foolishly waste their money sending datacenters to space.

he is trying to say that you didn't make Jev at all. you fine tuned a model for a particular domain while Jev works across all domains. seems different right?

it is legal if you can do it, doesn't mean they have to like it. keeping features for logged in seems super reasonable to me.

proof of not ai

i don't think there is a page on x that just gives you a login wall that a normal non-logged in user would go to. if you link to something like a tweet or a profile page it shows you them. these services are for scraping.

this entire thing is about people that aren't even users of the service.

JetKVM mini is ~$40 or ~$100 for 3.

All babies were exclusively breastfed in 3064 BCE.

I think graph being log makes it a little hard to understand that the median birth year was around 2000 years ago. On the graph that is in the flat part but the flat part was a really, really long time.

Love seeing people experiment with stuff like YQL (https://en.wikipedia.org/wiki/Yahoo_Query_Language) taking advantage of all the AI stuff definitely increases the power.

based on every country that didn't adopt capitalism until recently, absolutely not. there are very few non-capitalist countries as all of them either failed or changed their economies.

i wouldn't characterize it as very expensive or even expensive. it is in fact super cheap. even in italy.

TikTok does this so they can reuse popular content to game engagement.

this is within 500m of the datacenter. this is a deeply unserious problem. parking lots are probably a bigger contributor.

the energy consumption of EVs is already massively below gas vehicles. also, as others have pointed out, their tires already are within the regulations.

since this is about fuel consumption I assume that it doesn't apply to EVs

you are silly if you think this is limited to claude models

This custom prompt on claude.ai fixes it:

Use Simplified Technical English rather than being overly verbose.

You have direct control over whether he is in your feed.

No one produces a feed that purposefully delivers content or ads that you aren't interested in. Those are bugs. Go look at the published X algorithm. You won't find anything in there that would attempt to show you things that don't match your interests.

Algorithmic feeds are just search algorithms based on what the user is asking to see with their behavior.

w/e, plenty of prior art for this

I now have a fork of Ghostty that is kept up to date with AI and automatically triages and then PRs issues. All because I wanted to have a built in settings page instead of a config file I needed to edit. The future of software is custom per user.

A US gallon of deuterium water costs roughly $2,600 to $3,800, while a gallon of pure tritium water (super-heavy water) would cost about $44 million.

It is much harder to prove these conjectures true than false. On some of these people had spent years of their life trying to prove them true. By showing they are definitely false that can be avoided.

They are all different problems for the different languages. I was hoping this was a benchmark that attempted to see which languages were more efficient to use with which models.

seems very reasonable to me

What I have been seeing recently is having a great LLM rewrite the library leads to fewer bugs and also optimized for your use case. More expensive for sure than pulling something like epub.js but when the library has infinite open issues it can be a lot better.

Funnier than the article.