Hacker Newsnew | past | comments | ask | show | jobs | submit | Surac's commentslogin

where is my minority report?

this is a litte bit like the AI bubble works. I can't point to the thing with my finger but it feels wrong.

i realy hate it when well knowen world like "temperature" are missused to discribe something complete out of context. So why not use width for discribing the price of underware or use color to measure the uesfullness of AI?

It's not new, been used like that since the 80's. It scales the logits in a sum of exponentials.

so where are the millions of line code you need to train the LLM in your new language? Remember AI is just a statistic prediction thing. No input -> No output

intresting project, do you mind to explain what brought you to do that research? im a litte surprised that the more simple languages tend to use more tokens, but after thing i realizend that languages with more expressiv syntax allow to write with less "words". But i also think it is a little bit like a race of watches. who realy wants to know what watch runs faster?

Reinventing COBOL?

To bad I can not change away from vs code. Most chip manufacturers use the vscode + Addons approach today to give you a toolchain to Programm for your microcontroller. They (thank god) abandon there proprietary IDE attempts and concentrate on chips and compilers again. They found VSCode and now you are required to use this. I’m very unhappy with the situation and on the job hunt to get away from embedded to application programming

Which microcontrollers? Of interest, I program STM32, ESP32 (Risc-V) and Nordic chips, and do everything in RustRover. The compiling is decoupled from the editor. (You do cargo run to build and flash with debugging, or click the Run IDE button; works just like any program) The editor is for the code structure, and should reflect that, vs the chip. There is Cube IDE which is useful for configuring clocks, memory layout data, and other periph info as a supplement to the RMs and User Manuals, but it's not required to write code.

Do you do this in a professional setting? I'm curious because I did some embedded/uC work about 10 years ago and considering the state of C/C++ SDKs (and IDE support) at the time, I would have expected it to take decades for Rust to get a foothold.

Yep; do it at work (Security-related sensors for a DoD contractor) as well as my own small business. It doesn't have a foothold, and may not ever; we will see. I think a lot of the embedded rust content you see online is makers who are more interested in doing tricks with the ownership system and Async. So, I am an exception, but... I do recommend this workflow despite its lack of popularity!

I just like rust for the overall language and tooling. (For example, the workflow I described above); don't really care about the memory safety aspect to the degree it's often presented.

The biggest downside is I have to do a lot of leg work which wouldn't be required if done in C or C++. E.g. implementing hardware interfaces from datasheets and RMs. Sometimes there will be a Rust lib available, but in my experience they are rarely in a usable state.


As a hobbyist who's written and is working on a couple of async HALs my take is that Rust is well suited to embedded work but yeah there are hurdles. It's immature so while things like Embassy are a joy to work with, they're missing a lot of (sometimes seemingly basic) features.

Mostly Arm systems using CMSIS from TI.

So I think you and the person you responded two are talking about two different things: developing software with and without a HAL.

The rise in ARM brought about quite a bit of standardization. You're no longer bound to vendor specific compilers and toolchains. Insofar as you're willing to essentially reimplement large swaths of the HAL you're able to BYO dev environment. Of course all of this is also subject to the quality of the CMSIS packs and documentation put out by vendors.

This is true with Rust as well, and in this capacity Rust is quite mature and well supported for Cortex-M stuff (and to a slightly lesser extent Xtensa and RISC-V). The tools to create thin wrappers around the registers (so called Peripheral Access Crates — PACs) are pretty well fleshed out at this point.

If you're looking for a equivalent to first party HAL to leverage (e.g. CubeMX, Atmel Studio), Rust is significantly less mature here if only because of its age. In Rust land there are multiple different HAL frameworks to work with and it's likely you'd need to use a combination of them. Embassy (a combination of an async framework and HAL components) is pretty slick if it does what you need.


i think i might be brain damaged because i actually like the eclipse ides they put out, atleast the stm32 one

I very much recommend smt32cubeMX.

Lol here I am moving them on my own when vendors won't or do it wrong. Sad to see a fellow embedded leaving. We certainly need all the help we can get

Maybe he doesn’t like to be the next Maduro dragged out of bed at night by marines. So he complies to everything Trump wants him to say?

I think all the icon „bloat“ is from the designers that needed to stay relevant. They learned to use pictures instead of text from some professors and wanted to show how much they are worth. The picture for text pattern is valid if you want to replace one or two words in a commercial but not if you have about 200 icons on your page. Also the trend to a flat black and white design makes the icons just to blend together

Ubislop, Ubislop never changes. Never trust a Ubislop


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

Search: