blahgeek

Seems lispy

Actually it already exists. My microwave have a button labeled “AI” which suppose to run the microwave until it determines the food is ready.

This page works beautifully in my iPhone. As I scroll down, the content slides and animates. I actually came here to say that I’m stunned that this effect can be good in mobile, only to find out your comment :D

What’s wrong about this?

I wonder if you copying the term here in HN (with the link to that site) would pollute the benchmark dataset…

The worst nightmare would be the maintainers in turn use large language model to review or apply these patches

I think it’s two fold: the evolution of hardware (GPUs) that give us enough compute power, and the invention of novel algorithms (transformer) that can effectively consume and understand all these data.

Same situation when two people named John Smith want the same john-smith.com domain now. You may argue that in the current situation, if they are from different countries like US and Canada, they would obtain john-smith.us and john-smith.ca respectively. But in the proposed situation, they can also get john-smith-us and john-smith-ca

Yet python has two ways of writing string literals (single quote and double quote)

Does this mean that, when the user clicks a "share to social.network" button in random.com, the browser would first request the full html page of social.network (in the background) only to parse the rel="share-url" link? Seems a bad design.

As someone who have only used Emacs and Vim in the past 10 years, I wish you are right. But according to my observation, 90% of those 38% of developers only use Vim in when they are sshing to the server to update few config files or make simple edits to the scripts. When they do proper programming (like hundreds lines of coding in a project), they switch to other IDEs like VSCode. So yes I personally still consider Vim and Emacs “niche” editors.

Like LSP and tree-sitter, I think AI coding tools like Claude Code or Aider are very good news for niche editors like Emacs or Vim. Instead of struggling about implementing advanced IDE-like features, they can integrate with these tools relatively easily, and focus on other editing related features that set them apart. In fact, IMO it makes these editors more competitive because they are highly customizable and easier to integrate with these tools.

This makes more sense than it looks. semver is a lie because every change is a breaking change (Hyrum’s law)

Kimi.com

> Code is expensive in terms of tokens and harder to constrain in environments

It's also true for human. But then we invented functions / libraries / modules

I've been coding in Python for 10+ years but I can never really get python's tooling ecosystem. It seems that there's always a newer shiny choice. easy_install, pip, conda, virtualenv, pipenv, setup_tools, hatchling, setuptools-scm, uv, requirements.txt, pyproject.toml...

I wish python can provide an "official" solution to each problem (like in rust, there's cargo, end of story), or at lease, an official document describing the current best practice to do things.

Apparently Kagi has been down globally for >10 minutes now. I cannot believe such maintenance is allowed for a service that want to convince people as a Google replacement.

Given the title, I'm surprised (in a good way) that it's not about AI coding.

Asking it to explain rust borrow checker is one of the worst examples to demonstrate its ability to read code. There are piles of that in its training data.

And the bash script would be shorter and easier to write than the prompt.

I’ve been using vim for 10+ years. However I honestly don’t see the downside of repeating h or j to move up/down (with the key repeat delay adjusted to a small value). It’s more intuitive than using say 15j, which involves recognizing some number in the screen and then look at the keyboard to type because the upper number row cannot be easily touch typed

I think it's more common because one doing only coding can get paid reasonably. On the contrary, few people who "bake cakes for fun, make music for fun, write poems, novels, play chess for fun, practice sports, grow potatos" can get paid enough for a living, so that's usually not an option to consider. (Which is the reason that I find us coding people very lucky.)

> What I mean by doing research is not just Google it but also asking questions, calling people and using other resources.

> It’s obvious that currently none of the SOTA models can do such tasks, agentic or not. And therefore they are NOT AGI to me.

I myself almost never do that (calling people when googling is possible). Guess I'm not general intelligence. :)

> An often overlooked feature of the Gemini models is that they can write and execute Python code directly via their API.

Could you elaborate? I thought function calling is a common feature among models from different providers

The title seems to suggest it's a bug of the linux kernel, but it's not. The layout of an internal struct is never part of the user-facing API of kernel, so it's bound to be changing. eBPF programs are expected to be compiled against the exact kernel config that is going to be executed, just like any kernel modules.

> Isn't there basically Google/Waymo and then, seemingly much further behind, Tesla Cybertaxi, Amazon/Zoox, and Uber/Yandex?

Global-wise, there are also a few Chinese companies in the robotaxi market. Pony.ai and WeRide both recently went in public market.

Reading half way through the post, I thought what the author going to do was to analyze the distribution of numbers in the filename, and, I don't know, maybe give an estimation about how often people take photos or videos, based on the time, country, etc. That would be an interesting study.

The logic is sound. Storage can be end-to-end encrypted, but it cannot be used for computation.

I'm not impressed. I can totally imagine single person building the app under 5 months. Well, of course, it will not have the best UI or recommendation algorithm or marketing stuff, but I don't think those are in the critical path on the building of the app. I expected that a company like Meta can build apps like this in weeks.

It got me thinking that, in terms of building software 0 to 1, maybe the scalability of head count is surprisingly low.

That “python float” example is a hall of shame for Google, not python. I searched the same term using kagi.com and first results are exactly what the author expects.

Link: https://kagi.com/search?q=python+float&r=us&sh=cDUw0QAZf3DUs...