Compilation is CPU bound and, depending on what language mostly single core per compilation unit (I.e. in LLVM that’s roughly per file, but incremental compilations will probably only touch a file or two at a time, so the highest benefit will be from higher single core clock speed, not higher core count). So you want to focus on higher clock speed CPUs.
Also, high speed disks (NVME or at least a regular SSD) gives you performance gains for larger codebases.
I think the main barriers are context length (useful context. GPT-4o has “128k context” but it’s mostly sensitive to the beginning and end of the context and blurry in the middle. This is consistent with other LLMs), and just data not really existing. How many large scale, well written, well maintained projects are really out there? Orders of magnitude less than there are examples of “how to split a string in bash” or “how to set up validation in spring boot”. We might “get there”, but it’ll take a whole lot of well written projects first, written by real humans, maybe with the help of AI here and there. Unless, that is, we build it with the ability to somehow learn and understand faster than humans.
People seem to disagree but I like this. This is AI code used responsibly. You’re using it to do more, without outsourcing all your work to it and you’re actively still trying to learn as you go. You may not be “good at coding” right now but with that mindset you’ll progress fast.
We declare children as dependents legally, don’t we?
Nah I’m an innovator! I’ll just innovate a better chip that’ll never fail and software that has no bugs!
Proceeds to put Linux on a common SoC and load it with shoddy software from a low paid contractor.
At least the source wasn’t a Rick roll
dd if=/dev/null of=/dev/eng0
Oops!
Convert the PWD value to use backslashes, too, for extra cursedness.
What is this magical dictionary? I’m almost afraid to ask but, won’t just about any dictionary do just fine?
If that’s your concern I personally find ZeroTier a lot simpler to set up securely. You basically can’t expose things to the public internet through it because it doesn’t even require you to forward ports or anything.
I’ve observed the same thing about YT music’s audio. It’s actually a bit frustrating because YT has the better quality, it’s louder too (Spotify app is strangely quiet in comparison), the algorithm is nicer, I actually even like the UI a little better. But the queue system sucks donkey balls, there’s no cross-system control, and no jam so I often go back to Spotify when with friends.
As someone who used to run a Plex server and a jellyfin server for myself (not at the same time) I’d have to agree with the sentiment. If I were trying to provide it for my less techy friends/family I’d go Jellyfin again. But for just me? Video files + samba fileshare all the way. Even lets me play the videos on my phone.
Well, not exactly. For example, for a game I was working on I asked an LLM for a mathematical formula to align 3D normals. Then I couldn’t decipher what it wrote so I just asked it to write the code for me to do it. I can understand it in its code form, and it slid into my game’s code just fine.
Yeah, it wasn’t seamless, but that’s the frustrating hype part of LLMs. They very much won’t replace an actual programmer. But for me, working as the sole developer who actually knows how to code but doesn’t know how to do much of the math a game requires? It’s a godsend. And I guess somewhere deep in some forum somebody’s written this exact formula as a code snippet, but I think it actually just converted the formula into code and that’s something quite useful.
I mean, I don’t think you and I disagree on the limits of LLMs here. Obviously that formula it pulled out was something published before, and of course I had to direct it. But it’s these emergent solutions you can draw out of it where I find the most use. But of course, you need to actually know what you’re doing both on the code side and when it comes to “talking” to the LLM, which is why it’s nowhere near useful enough to empower users to code anything with some level of complexity without a developer there to guide it.
You can get decent results from AI coding models, though…
…as long as somebody who actually knows how to program is directing it. Like if you tell it what inputs/outputs you want it can write a decent function - even going so far as to comment it along the way. I’ve gotten O1 to write some basic web apps with Node and HTML/CSS without having to hold its hand much. But we simply don’t have the training, resources, or data to get it to work on units larger than that. Ultimately it’d have to learn from large scale projects, and have the context size to be able to hold if not the entire project then significant chunks of it in context and that would require some very beefy hardware.
Ah yes the ever elusive “tech debt”
Thanks for the summary, I did a bit of reading myself. It’s interesting the dynamics at play here - you’ve got a long, long term contributor in Hellwig who’s been a maintainer since before Rust even existed, then you’ve got quite a few people championing Rust being introduced into the kernel. I feel like Hellwig’s concerns must have more to do with the long term sustainability of the Rust code - like will there be enough Rust developers 10, 20, 30 years down the line. I mean, even if it stays maintained, having multiple languages in a codebase increases complexity and makes it harder to contribute. Then you have Filho resigning from the Rust for Linux project, which in itself kind of calls into question the long term sustainability of the project. It seems like Rust would have quite a few benefits for the Linux kernel, but the question remains of if it’s still gonna be any good in a few decades. This is juicy stuff!
Anyone got more context on this I can read through? I haven’t kept up with this other than Linus’s notorious attitude.
I think mint is crazy better these days compared to 10 years ago, and it probably just came down to “we want to be user friendly to those who need their hands held” crashing into “actual users who need their hand held are trying it out.” 10 years ago, I think there simply wasn’t enough interested in Linux outside of Linux circles to properly test and figure things out, not to mention the strides the software itself has made in supporting more hardware more seamlessly.
The thing about RTFM is that users don’t, and the users that stuff like Mint is geared towards is those who when asked to read a wiki page, will simply give up. Windows has a cottage industry of people who do various things to make it easier for that kind of user. For example, just installing Windows on a device for you (albeit with bloatware usually) complete with all the drivers for your hardware. For most of the hardware on a laptop (audio, internet, HIDs, USB), that’ll have you set for life without having to touch anything and for the graphics that’ll at least have you set for several years without having to touch anything. And it’s not like Linux doesn’t have this level of support, it’s just that Windows has this level of support for consumers and Linux typically has it relegated to the enterprise sphere.
That being said, it’s insane how easy it is now to just install Mint, or PopOS, or even Ubuntu and have a working system. But most users don’t even install their Windows, much less a completely foreign OS.
It’s remarkable, really.
You should give Claude Code a shot if you have a Claude subscription. I’d say this is where AI actually does a decent job: picking up human slack, under supervision, not replacing humans at anything. AI tools won’t suddenly be productive enough to employ, but I as a professional can use it to accelerate my own workflow. It’s actually where the risk of them taking jobs is real: for example, instead of 10 support people you can have 2 who just supervise the responses of an AI.
But of course, the Devil’s in the detail. The only reason this is cost effective is because of VC money subsidizing and hiding the real cost of running these models.