> It’s definitely a better Google for most searches
I can't agree with this. You tend to get one point of view, often without any actual resources and references so you have to go look it up yourself, on [insert search engine]. Plus, what does it say when we consider an AI the one stop for our data intakes.
One of the main (vocal) issues people seem to have with IPv6 is that the addresses are hard to remember. But having eight different three digit numbers (r.r.r.r.n.n.n.n) does not seem any easier unfortunately.
I would think that we are going to see RAM prices increase even more, given, among other things, pure helium disruptions and increased electricity prices.
I haven't looked closely into TurboQuant, but perhaps it will revolutionize just as much as the 1-bit llm did...
I admire this mindset, and this is what I try and do as well with my projects.
But for larger projects, on which the giants rests, (I'm thinking cURL, ffmpeg etc.) it most likely stops becoming/feeling like charity. Especially, since a lot of people do not see it as charity, and thus tries to force the maintainer(s) to do even more unpaid work.
> 82 percent of Maven Central demand comes from fewer than 1 percent of IPs
This is interesting, and the author goes on to say that 80% comes from the largest cloud providers. But I wonder how much of that is coming from CI pipelines and how much is actually the cloud provider's usage?
I agree, I like some of the directions the fork would go and dislike some. The apparent, fork, publish on HN, then change (and the change showing not a lot of understanding) makes me throughly question the legitimacy and long term stability of it.
I don't think it is too bad, the telemetry it sends is quite rudimentary. However, would have been a good move from astral-sh to be open and explicit about it, and allow turning it off.
Out of curiosity, what are you proposing instead? I'm currently working in a small company (less than 4 full time employees). And while we can never support native apps for all the platforms I have been wondering what we would use instead of a web app?
The advantage of the web app is that it just works, without installation, so there's no friction there. I'd very much prefer a native app, but the overhead is quite high, no?
Recently I had to use Julia and their Pluto notebooks. While I really did not enjoy Julia, I loved Pluto notebooks, and Marimo is basically this, but for python.
The reactive component is great for exploring data in an easy way, with sliders and buttons. Also for simulations when you want to try a bunch of different parameters.