ogogmad

39 ms·

691.
by ogogmad 4y ago
"Kin"? If you mean to be anti-Semitic, you'll find yourself banned from this website.
692.
by ogogmad 4y ago
"Bare minimum" is certainly an exaggeration, so I apologise. Gnome is indeed perfectly usable out of the box. And KDE is not "far more usable", it's just more convenient in lots of small ways. I haven't used it
693.
by ogogmad 4y ago
Even if it's "bloated", I've found KDE to be a far more usable DE than Gnome. KDE provides you with more features than you need (or thought you might need), but everything's made discoverable, so you don't mind
694.
by ogogmad 4y ago
The choice stuff might make vector and GPU programming more natural, assuming it can be made efficient enough. The syntax it provides is certainly convenient for that. Might even be an advance over array programming in some ways, by being m
695.
by ogogmad 4y ago
Sign up and try it. Highly recommend.
696.
by ogogmad 4y ago
The following is snarky, but since it got published in The Atlantic (-- a well known publication?), what does it mean? Proponents of LLM generativity may brush off this concern. Some will do so by glorifying GPT’s obvious and fully realize
697.
ChatGPT Is Dumber Than You Think(theatlantic.com)
5 points by ogogmad 4y ago|4 comments
698.
by ogogmad 4y ago
If you allow LLMs (Large Language Models) to directly call out to APIs, like POSIX, Vulkan, Win32, X11, etc. then it may become possible to program things using a "Socratic" method that involves producing no code at all. You simpl
699.
by ogogmad 4y ago
I agree. It's a "live and let live" movement. Not some form of austere modernism. Beauty of people is indeed subjective. Different ethnicities have different innate preferences.
700.
by ogogmad 4y ago
[edit: Was snark. Sorry. I also may have misunderstood the parent comment.]
701.
by ogogmad 4y ago
That wasn't Hofstadter's point. Hofstadter didn't try to "disprove" the possibility of having AI -- he only set out to show that certain approaches couldn't work. GPT's approach isn't one of those he
702.
by ogogmad 4y ago
Introducing: AI-Oriented Programming. Coming soon. If you allow a tool like ChatGPT access to APIs, I think it opens up a new programming paradigm: AI-Oriented Programming. The idea's that you talk to it, and tell it what to do in resp
703.
by ogogmad 4y ago
I think people don't really understand what's happening, and therefore why it's impressive that it's happening.
704.
by ogogmad 4y ago
I don't know about rollups stuff, but Lightning Network-type 2nd-layers are prone to flood and loot attacks. I don't see a way to make them secure. I'm happy to be enlightened.
705.
by ogogmad 4y ago
I once asked an earlier version of GPT a question that it was never asked before, and it will never be asked again, and it gave multiple imaginative and plausible answers to it. It's not a bullshit machine.
706.
by ogogmad 4y ago
It understands the "pseudo" in pseudorandom. (That is, the determinism). Pseudo + "random". [edit] The application is it might be able to denoise text files full of "noise".
707.
by ogogmad 4y ago
If it walks like a duck and talks like a duck, it's a duck. > They are big tables of tokens with probabilities And we're just a bunch of carbon atoms. Let yourself feel wonder.
708.
by ogogmad 4y ago
Somebody should run Ed inside of it; a text editor absolutely made for this. Have we finally found an application for Ed in the year 2022?
709.
by ogogmad 4y ago
It suggests that it understand what pseudorandomness looks like. A hash function output usually looks pseudorandom. It's almost like it can look at a picture from an old television, and separate the random static from the "signal&
710.
by ogogmad 4y ago
https://news.ycombinator.com/item?id=33842590 Also, there is a version of the law of sines for tetrahedra and higher-order simplexes, similar to the one the AI tried to use. It's an imaginative wrong answer.
711.
by ogogmad 4y ago
British English has a very flat hierarchy, with no T/V distinction as found in French or Russian, and very little use of honorifics or titles (even for authority figures like university professors). British politeness is not based on h
712.
by ogogmad 4y ago
The first example is trivial, and serves as an illustrative but useless example (which I don't need). I don't know enough about the second example.
713.
by ogogmad 4y ago
But why do you need category theory to understand lists and monoids? They're so basic and elementary.
714.
by ogogmad 4y ago
I think it's just compression: - An operation may be "functorial", meaning that it preserves more structure than perhaps originally thought. For instance, the "fundamental group" operation is indeed functorial, whic
715.
by ogogmad 4y ago
Where do adjoint functors occur in CS? They occur in advanced algebra, and they occur in topology, but where else? And indeed, the fact that they preserve limits/colimits may help speed up communication and thinking. But I'm not s
716.
by ogogmad 4y ago
This is coming from the opposite extreme. In constructive mathematics, abandoning Countable Choice may allow: - The complex numbers to not be closed under square roots. - Space-filling curves to no longer exist. And subtle complications aro
717.
Logic and Linear Algebra: An Introduction(arxiv.org)
1 points by ogogmad 4y ago|0 comments
718.
by ogogmad 4y ago
> second by detracting by spreading the impression amongst the general population that experts aren't really experts How do we know who's an expert here? Peer review is only as good as your peers. Genuine question. It looks lik
719.
by ogogmad 4y ago
It looks like steady progress is being made in Mathematical Physics (which I believe is physics done from rigorous mathematical definitions - so you can no longer bullshit the Dirac delta "function" for instance). Some of that stu
720.
by ogogmad 4y ago
Relevant: The Counterfeiters. https://en.wikipedia.org/wiki/The_Counterfeiters_(2007_film)

More ›