wtetzner
wtetzner on Hacker News
> Why do people keep getting this wrong?
It's kinda funny to say people are getting it "wrong" when strong/weak typing isn't really defined anywhere or universally agreed upon.
In fact, even "dynamic typing" can be considered to be an oxymoron, since types are formally defined to be a static property of a language. A more correct term would be "uni-typed".
I'm not saying that the meanings you're using for static/dynamic and weak/strong are bad, just that chastising someone for getting them "wrong l" doesn't make sense for terms that are not widely agreed upon.
They definitely have other intentions, but it's often the same as people contributing in their free time: they use the software, want to make it better for their purposes, but don't want to be the ones to maintain it.
Of course not? I don't think anyone claimed it was out of charity.
Sure, but why not have a better virtual keyboard in the Apple TV OS?
Sure, I guess, but I've yet to see an LLM generate code that I'd consider either easy to read or particularly well-written, regardless of who was driving the agent.
And even if it does end up being good code, if it's something non-trivial, having a reasonably deep understanding of what it's actually doing is quite hard simply from reviewing the code.
I find using lifetimes to be simpler and easier to follow than reference counting, unless you're in a situation where you actually need a shared lifetime.
Adding Rc/Arc is just another layer of indirection and noise that makes the code hard to read.
Honestly I think having blanket rules in either direction is a bad idea. Sometimes it makes sense to qualify things, sometimes it's clearer if you don't.
> Of course shareholders praise other things like cutting costs and raising revenue.
And I don't think these software releases really have anything to do with that. I would be very surprised if they were meaningfully driving sales.
It's funny how differently people see things. I find reading/reviewing LLM generated code to be much more difficult and tedious that writing it myself.
In my experience LLMs tend to do a very poor job at architecture.
The way things have been going, it seems unlikely we'll see a price drop any time soon.
Neither of those things make it any less ridiculous. What it really amounts to is that you're not really purchasing the hardware.
> If Sony and MSFT decide to go down this route and it's the only way to play blockbuster games, it'll happen whether we like it or not because the market cares far more about playing the latest game than it does maintaining control over its own destiny so to speak.
Maybe, or maybe it'll live a giant gap waiting to be filled in by a new competitor in the console market.
Just because something is due to incompetence instead of malice doesn't mean they shouldn't be held accountable.
You don't store your password in the URL.
Couldn't you distinguish things by their grayscale value? Then you change the hue on top of that for people who can see color.
On my phone I was able to just touch-and-hold on the screen to move the camera. I assume on a PC you can just use the mouse to control the camera.
Then why not just ask the LLM to explain what LSP stands for?
It's not just habits. It changes the composition of your gut microbiome, hormones, etc. All of that can make changing your diet much harder than one might think.
This sounded weird to me at first until I realized that this exactly what I do when watching a movie with subtitles.
I'm wondering if going back to cheaper and less capable hardware will actually result in better games. Forcing creativity, and maybe (by necessity) dropping the constant push for higher fidelity graphics. I think that push for fidelity is so costly that a large number of game developers don't even attempt to innovate in the rest of game design anymore.
> For the ps6, they were already done with design and they'd need to let that go to waste to not put it out.
I guess nobody can predict the future, but this sounds like the sunk cost fallacy. E.g., if it's not going to sell well enough to recoup manufacturing costs, then producing the console will be more costly than just wasting the design effort anyway.
Maybe they don't need mass market sales to at least break even?
> The author knows the orientation of his language better than anyone else.
^ I was only addressing this comment, not the specifics of Odin.
> Odin is a general purpose language; is capable of being used in numerous different areas from application development, servers, graphics, games, kernels, CLI/TUIs, etc.
I would argue that being capable of doing anything is not in conflict with being oriented around something specific. Any Turing complete language is capable of those things.
Again, I don't even know Odin, so I'm not addressing whether or not it is game dev oriented. Only that the arguments themselves don't seem sound.
EDIT: grammar/formatting
> that seems backwards to me
I'm not sure I completely understand which part you find backwards. Do you mean merging language runtimes and the OS is a bad idea, or that you think merging them would lead to more innovation?
I can see an argument for both (in terms of innovation), but being able to run only one language environment on a computer at a given time would make it much harder and heavy weight to use new languages. Or at the very least, new language runtimes.
Yeah, I kinda feel like when you're mentoring people, you should let them drive. If you're going to do it for them, you might as well just record a training video.
This was already addressed:
> But with LLMs generating 7101 commits in less than a month that isn't feasible.
I don't think trying to understand LLM generated code is feasible for anything other than very small projects. IMO it's a big problem with using LLMs for coding. Sure, they can generate a bunch of stuff, but in some ways that just makes the real problems of software development even harder.
I kinda feel like using memorization techniques for things that require deeper understanding probably isn't the most efficient way to learn.
IMO you want to be actively trying to map the new concepts to things you already understand, and constantly working to update your mental model.
So your solution is to just read through all 1M+ lines of code?
What about just not installing sketchy extensions?