For the best experience on desktop, install the Chrome extension to track your reading on news.ycombinator.com
Hacker Newsnew | past | comments | ask | show | jobs | submit | history | more smartmic's commentsregister

A question I only dare to ask myself in these times of LLM: Is this even a real human being or already an instance of an ‘agentic system’?


Agreed. I once used it for data preparation for a data science project (GNU APL). After a steep learning curve, it felt very much like writing math formulas — it was fun and concise, and I liked it very much. However, it has zero adoption in today's data science landscape. Sharing your work is basically impossible. If you're doing something just for yourself, though, I would probably give it a chance again.


I actually confused AsciiMath with UnicodeMath[0]. Interesting that this hasn't been mentioned here yet. Check out its playground[1].

[0]: https://www.unicode.org/notes/tn28/UTN28-PlainTextMath-v3.1....

[1]: https://murrayiii.github.io/UnicodeMathML/playground/


It would be much more readable if AsciiMath[0] is used and still gives you the benefit to render it with MathJax if required.

[0] https://asciimath.org/


There are also ASCII-art ways of writing formulas. An LLM should be able to produce these.


I use XQuery to transform XML data. Whatever can be done with XSLT can also be done in XQuery. It is a functional language and, unlike XSLT, it is supported by more tools for newer versions (XSLT > 1.0 is only supported by Saxon, as far as I know). Overall, it feels much more modern and ergonomic for querying and transforming XML. Best of all: XQuery 3.1 supports JSON natively — I have also adopted it for JSON in some ETL pipelines.

Check Wikipedia for more information: https://en.wikipedia.org/wiki/XQuery


It would be wonderful if this were the final blow for Windows as a classic PC operating system and if Linux were to establish itself on the desktop as a more than reasonable, indeed unavoidable, alternative. The problem with AI is that it offers apparent solutions to non-existent problems. How long will big tech continue to back a horse that the general public does not want to ride?


There's sadly whole business and services sector that relies on deals with MS.

My mother mid pandemic placed an order for a hearing aid which is partially supported with money from our healthcare system. Unlike 10-15 years ago nowadays everything is digitalized and done by the Internet - clients are no longer running around the city with paper forms. Sadly that means every personal information, data is exposed to Windows and with W11 it's even less possible to avoid being harvested. Not mention always online browser software used for hearing tests that surely collects on its own.


> The problem with AI is that it offers apparent solutions to non-existent problems

Yes

The good thing about AI is it offers assistance with old, tractable, problems

When it is everywhere it is an unhelpful annoyance


> The good thing about AI is it offers assistance with old, tractable, problems

That doesn't sound very good. I'm not sure if it's sarcasm or a typo. If the problem is old and tractable, you don't need AI for it. You could use a simpler, more efficient, and more reliable technology.


Good point!

"old, tractable" and hard problems.

Routine things.


PC manufacturers are bribed to use Windows, as always. Windows has a Soviet problem, it's like an Iron Curtain and most users will never thing of leaving it, if they do it's with kgb tourism agent called WSL. If Windows is a horse your certainly do not ride it, it rides you.


Only in my dreams.


Winter is a cyclical concept, just like all the other seasons. It will be no different here; the pendulum swings back and forth. The unknown factor is the length of the cycle.


Here is more information from German newspaper Sueddeutsche Zeitung[0] (use your browsers language translation feature), some excerpts:

> In Germany, however, such images—US soldiers at food banks—are unlikely to be seen. The list of supply tips quickly disappeared from the website. Why food banks were mentioned there in the first place remained unclear at first.

> The more than 30,000 US soldiers in Germany are unlikely to have encountered any major financial problems anyway – according to the IMCOM spokesperson, they received their October salaries as normal despite the shutdown.

> This is in contrast to the approximately 11,000 civilian employees of the US Army in Germany. The US has not released their pay, which is processed by the German “Wage Office for Foreign Armed Forces” in Rhineland-Palatinate. For the time being, the [German] federal government has stepped in and paid the logisticians, cooks, and firefighters their October salaries totaling around 43 million euros, as a spokeswoman for the [German] Federal Ministry of Finance announced on Thursday in response to a query from the SZ.

[0]: https://www.sueddeutsche.de/bayern/deutschland-us-army-solda...


The reason I was put off by Rust was compiling from source. I experimented with a ports collection package management system similar to those used in BSD a while ago, and every time a Rust program needed to be compiled, I could go to sleep; no, basically rendering the system unusable. It was like the dependency abyss of NPM combined with the worst possible compile times, even worse than C++.


Even worse than C++? That has never been my experience. I maintain C++ projects that take hours to build from scratch. Most Rust projects I install, even the very big ones, are ~3-5min max.


Are you building in release or debug? Do you have lto enabled? Rust compile time is absolutely worse than c++, and I'm not sure many in the community would disagree. You need to aggressively split large crates up to make it sane. A lot of this comes down to the fact the crate is the compilation unit rather than individual files.


It would appear that when actually measured head to head, compile times are approximately the same: https://quick-lint-js.com/blog/cpp-vs-rust-build-times/


For CUA aficionados, I recommend dte[0]. It has replaced my nano usage quite a lot.

[0]: https://craigbarnes.gitlab.io/dte/index.html


I personally like msedit.

https://github.com/microsoft/edit


Promising but very barebones. Hard to do without syntax highlighting these days, for example. But I think it would be useful on a tiny machine with openwrt, as micro is huge there. ;-)


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search:

HN For You