vrighter

34 ms·

661.
by vrighter 1y ago
developing a ui != developing an os.
662.
by vrighter 1y ago
are you training a new model or is this just prompt "engineering" with a fancy ui?
663.
by vrighter 1y ago
AI in notepad is not something that should be in an OS. Any OS. The os is supposed to be there to sit between your applications and your hardware. It's supposed to offer some sort of interface for you to operate your machine. Having ar
664.
by vrighter 1y ago
AND THIS CODE IS GARBAGE!
665.
by vrighter 1y ago
not american, but we had decent ev grants here. Never really factored them in. I needed to buy a new car one day, and the only small one i could find was electric. The grant on top of that was just a happy bonus, i would have bought it anyw
666.
by vrighter 1y ago
the mostly-er it works, the harder it is to find out which tiny part does not work.
667.
by vrighter 1y ago
if an llm is unreliable, then why would another just-as-unreliable llm make it any better?
668.
by vrighter 1y ago
if additional sensors improve the ai, then your last statement is categorically untrue. The reason it worked better is that those additional sensors gave it information that wac not available in the video stream
669.
by vrighter 1y ago
which cameras have stereoscopic vision and the dynamic range of an eye? Even if what you're saying is true, which it's not, cameras are so inferior to eyes it's not even funny
670.
by vrighter 1y ago
The reason it gets very difficult to distinguish between the two is that there is nothing to distinguish between the two other than subjective human judgement. When you try to be objective about it, it's some input, going through the s
671.
by vrighter 1y ago
Same here. My favourite of those is "none of them even being present on my PC"
672.
by vrighter 1y ago
Yeah, that does not mean type erasure. There were just no generics. That is just making use of polymorphism in the language. At no point is there a difference between what the language knows a container can contain, and what the runtime kno
673.
by vrighter 1y ago
That is just not a solid argument. There are countless examples of LLMs splitting "blueberry" into "b l u e b e r r y", which would contain one token per letter. And then they still manage to get it wrong. Your argument
674.
by vrighter 1y ago
no, they can't. That's been tried before. The result was windows 8.
675.
by vrighter 1y ago
throwing a pair of dice and getting exactly 2 can also happen on the first try. Doesn't mean the dice are a 1+1 calculating machine
676.
by vrighter 1y ago
no, when the signal first goes high, the button is immediately considered pressed. The debouncing stops it filters out several transitions after, so that the button stays pressed until the circuit has settled in the low state.
677.
by vrighter 1y ago
so? doesn't change the fact that it fits the formal definition. Just because llm companies have fooled a bunch of people that they are different, doesn't make it true. If they were different things (objectively, not "in my op
678.
by vrighter 1y ago
How is "this sentence is in first person" when the sentence is actually in third person not a hallucination? In a question with a binary answer, this is literally as wrong as it could possibly get. You must be doing a lot of menta
679.
by vrighter 1y ago
For starters, if it were superintelligent it would eventually make discoveries. New discoveries were not in the training set originally. The model needs to be trained to use the new discovery to aid it in the future. As it is, it has to kee
680.
by vrighter 1y ago
they are improving exponentially... but the exponent is less than 1...
681.
by vrighter 1y ago
but we have green belts now!
682.
by vrighter 1y ago
As soon as you changed any gate layouts, it stopped being a gate level emulation of a 4004. It's just a 4004 emulator (with extra steps). The gate level simulation is of a different device that emulates a 4004. To be fair, the whole
683.
by vrighter 1y ago
When did c# have type erasure?
684.
by vrighter 1y ago
The most striking thing about this project is the insane number of ignored issues asking for the training code.
685.
by vrighter 1y ago
I actually really like the AmigaOS way. Where all windows have this behaviour. In fact, bringing something to the foreground is an explicit action. Sounds weird, but it allows you to, for example, drag something from a background window to
686.
by vrighter 1y ago
Great. It's really annoying to be typing my password to unlock my password manager, immediately after logging in, only for discord to pop up and steal focus, because it takes long to load. Switch back, try entering the password again a
687.
by vrighter 1y ago
To any third party that claims to need to come between me and any of my auth keys, I have only one thing to say: "Fuck off".
688.
by vrighter 1y ago
Which models release the full training dataset, the training code, the inference code, and any random seeds used? Without all of those, you can't "compile" the project. If you can't compile it yourself, then it's no
689.
by vrighter 1y ago
What is it? The page goes on and on about how compatible it is but doesn't once mention what the thing actually is .
690.
by vrighter 1y ago
other post: https://news.ycombinator.com/item?id=44692665

More ›