Not surprised about Python, since it became the go-to language for 'everything AI' related even before the huge boom. If you ask to write a simple script in Claude Code, it will, in my experience, write it in Python.
OpenAI buying Astral and Anthropic buying Bun shows that the top 2 lines will grow even more, in my opinion.
Public repositories are skewed highly by vibe coding and automatic code generation, again in my opinion, so it also depends on what the clients think about their 'default' language to generate code.
I still remember that impulse, before AI, of googling everything before thinking a bit. AI as a technology is incredibly new and we're still figuring out how to use it, but things will stabilize. 'googling' everything also didn't take away our ability to think, so as long as you're aware, you can do something about it.
> A Gallup survey released Wednesday shows 71% of Americans are somewhat or strongly opposed to the building of AI data centers in their area, mostly out of concern for their environmental impact.
I know that nuclear power plants are getting safer and safer. but saying that this is the main concern..
> 5 million gallons. That’s how much water large data centers can consume per day, equivalent to the water use of a town with a population of 10,000 to 50,000.
Quick search online tells me that a nuclear power plant uses as much as 35–65 million gallons...
All of this apart from the actual danger of having a NUCLEAR power plant near home.. The accidents in the past were rare but the ones that existed we still mention them by name with the impact well know.
They can be built with so called 'dry-cooling towers'. It's just rarely done, except for a few in very arid regions so far.
Because it is more expensive initially. Considering the expected lifetime of about 60 years OTOH, this looks like 'peanuts', compared to the other operating costs, and indeed saving some because of much less used water over time.
How's the Object Storage support? did you have any issues? Hetzner looks more stable now but I had a lot of issues with S3 compatible storage a while ago... Almost made me change everything. If you're willing or able to share of course..
Design is amazing. Well done.
I inserted my birthday and got "Enter a real Gregorian date.".
Only then I understood I needed to insert "mm dd yyyy". User error for sure :) but this date format really annoys me
> Long running work: an agent doing a 10 minute task isn’t a ‘request’, it’s a long-running async process.
Correct, but we solved this a long time ago when we started sending files to servers to be converted, for example. We either got a 'job_id' or a call to a webhook when the job was finished."
The article argues that although this is an asynchronous process:
> Long running work: an agent doing a 10 minute task isn’t a ‘request’, it’s a long-running async process.
it should not be stateful at the database/storage level:
> Stateful compute: an agent might run multiple turns of a conversation, might process multiple tool calls, and relies on accumulated context. That state is not really ‘database state’, it’s the agents memory.
According to the author, the problem is already solved, but implemented with the wrong design assumptions.
(Uploading a file for conversion could be framed as a slightly different problem from the author's, though, due to size constraints)
Indeed. I'm wondering if Apple's "miss the train" with AI ended up being a blessing for them. Not only in the Google deal but also there's a lot of people doing interesting stuff locally..
