wasting_time

From the interview with an Emacs enthusiast: "In Jupyter you modify the state of the program. In Emacs, you modify the state of the art." :D

There's also "How to type “blimpy” in Emacs": https://news.ycombinator.com/item?id=37366341

So, how are people actually using Pi? Here I am with Claude Code and Codex in a terminal like a caveman.

Ah, I get it now, thanks!

The model is not available yet, so Google is essentially saying "trust me bro".

For anyone curious why they should choose one over the other:

https://forgejo.org/compare-to-gitea/

I don't get it. Can someone explain?

I think you can do this with Jitsi: https://jitsi.org/

They are having a massive surge because Claude is down.

The full interview was more interesting than this piece.

https://fossforce.com/2026/08/the-foss-force-interview-with-...

It will be interesting to see if LLM can push beyond the intelligence embedded in language.

General intelligence may still be a ways off.

Please don't use LLM to write HN comments. To quote the guidelines[0], "HN is for conversation between humans".

[0] https://news.ycombinator.com/newsguidelines.html

Off-topic, but clicking this link instantly crashes my desktop in both Firefox and Chromium. Debian 13 with backports kernel, AMD laptop graphics.

I have never seen anything like it in 20 years of Linux desktop use.

Evry, not Every.

Does this mean exe.dev will become open source? Or is it not a developer tool?

It's a service, and I want to add a CLI for humans and agents. Aeolos' reply makes a lot of sense, I can refactor the backend into a library and add thin UI and API routes on top of it.

You sort of answered your own question.

Another reason is compiling with -Werror and hitting new warnings.

Also, the version of the spec changes with almost every major release.

I made a UI with htmx, it works great.

But now I want to add a JSON API to the program: I could reuse the same routes (/create, etc), but require an application+json header; or add a full /api/ component essentially duplicating the logic from the main app.

Neither option is great.

I do the same for translated code. It's not creative work which is a prerequisite for being copyrightable.

And avoid relying on direct LLM output for actual work to make sure I don't accidentally include some regurgitated snippet from an incompatible license.

It helps that LLMs struggle to write good, idiomatic code in my language of choice.

The era of LLMs is similar to when Magic was discovered in the 1400s.

The layperson may be able to get ahold of a spellbook, but without Understanding it comes with high risk of turning your niece into a frog.

Whereas Wizards can cast increasingly powerful spells that build on each other, and make Art.

They may be thinking of Google's coarse location feature where the handset checks SSIDs in the proximity against Googles database for fast lookup.

Link for the lazy:

https://xkcd.com/1741/

git subtree was the OG tool: https://apenwarr.ca/log/20090430

It has since been merged into git proper:

https://manpages.debian.org/testing/git-man/git-subtree.1.en...

https://docs.github.com/en/get-started/using-git/about-git-s...

This site is unreadable on mobile: the left side menu takes up almost the entire screen and zooming in on the content does not work.

I wish NATO would step in and annex the Russian region between Ukraine and Kazakhstan. It could become a new country rebuilt from modern standards. And permanently cuts Russian supply lines to the middle east.

Losing territory is the only thing Putin will see as a loss. That would force him to admit failure.

No source code. Maybe I'm spoiled after decades on Linux, but I really don't want to hand over my phone to an opaque binary blob.

Relevant xkcd: https://xkcd.com/3230/

Whoops. My mistake for only reading HN headlines and extrapolating. OP, please disregard this recommendation.

Fair point. Probably I should assert copyright on the translated work. In any case it ought to be part of the GPL. The current text is rather ambiguous.

Look for a "static site generator". Bearblog and Hugo are popular ones. Then you can host your site anywhere and don't have to worry about security problems.

The GPL isn't strict enough now that everyone has a copyright laundering machine at their disposal.

I've been experimenting with adding a "translation" clause in accordance with the GPL paragraph 7, where if by means of automatically translating the project to a different language (by transpiling or machine learning) the license is retained, but not necessarily the copyright.

So far I'm not happy with the results by Claude and ChatGPT. Probably I should talk to a lawyer.

The GPLv3 was largely invented to prevent tivoization, we need GPLv4 now to prevent copyright laundering.