jiehong

When AOT came to be, I kinda hoped for GraalVM's solution would make it relatively as-is to open jdk.

Instead, this AOT feature is not very ergonomic to use at all, with too many pitfalls to be of broad use IMO.

That's really great! Congratulations to everyone working at Mistral!

Now, perhaps Kimi 3.1 is missing from the benchmarks, and Deepseek 4.1 Flash is still has impressive!

Agreed. I find Parakeet or Qwen ASR better.

I’m still sad Voxtral 3 wasn’t made open source by Mistral.

Perhaps we’ll reach a point where humans might have to record themselves writing the things and publish this next to the work itself to prove it was human made.

You just made me realise that latency might be a strong reason why some users might enjoy TUIs more.

I’m also thinking that there might be a universal solution to all animations: user-driven animation end. Like, the system could cut off the animation as soon as the user hits another key, making it faster and learning the right rate of animations over time. That would be great!

Yes, this only works if you can register more than 1 ssh keys on the services you want.

Nice!

Yet, with so many dictation apps, it makes me wonder how many are honeypots or something.

It’s more like devs only test on chrome.

Maybe the shell or the terminal itself could do that.

Some terminals already allow you to search they buffer after commands have executed, so maybe it’s a small step away from being able to use that as a cache.

How I would wish to have no overhead type aliases in Java…

I’m happy to see Elm evolving well like this!

Hopefully, someone can do some work on decreasing the requirements needed to build rust itself and ease bootstrapping.

We’re triangular plantations really interesting compared to grid? Unless you dealt with triangles of variable sizes (which sounds a pain to manage).

Contrast this with zig, where compiler performance is so important that it has influenced the language design.

Zig and Go are similar on that front.