rileyphone

11 ms·

61.
by rileyphone 3y ago
Note that Common Lisp contains CLOS, which is one of the most advanced object-oriented systems even now. Most Lisps are not functional like Haskell is.
62.
by rileyphone 3y ago
There's a decent amount of research that supports this view [1][2]. How else do you explain the unreasonable effectiveness of transformer-based language models? 1. https://journals.sagepub.com/doi/pdf/10.1177&
63.
by rileyphone 3y ago
Wow, that looks so bad from an anti-trust/competitiveness standpoint. M$ is embracing AI just like it embraced the internet 25 years ago.
64.
A Latent Space Theory for Emergent Abilities in Large Language Models(arxiv.org)
15 points by rileyphone 3y ago|4 comments
65.
by rileyphone 3y ago
What is the language you're working on now?
66.
by rileyphone 3y ago
I believe the web currently possesses the capabilities to make a networked Emacs/Smalltalk style system, souped up with AI, targeting web applications that don't require downloads; I've been pursuing such a system for a while
67.
by rileyphone 3y ago
Gates set up the (toxic) culture that continues today, and still provides high-level input. Definitely has something to do with their choices in the same way Jobs still does at Apple.
68.
by rileyphone 3y ago
Fun fact: WebObjects was the most expensive product Apple has ever sold, at $50000. It managed to keep NeXT alive after the hardware business collapsed
69.
by rileyphone 3y ago
Andres posted a reply on Twitter: https://twitter.com/algekalipso/status/1666091889500758016
70.
by rileyphone 3y ago
My preferred version of this: x.*(2).+(1).-(3).%(x) Unfortunately our brains are broken by pemdas and need clear delineations to not get confused; this syntax also extends to multiple arguments and is amenable to nesting.
71.
by rileyphone 3y ago
The bigger joke is their “open in app” links don’t work for me on mobile safari. At least old reddit still works on desktop.
72.
by rileyphone 3y ago
Given how PALM2 had less params than the first one (540B) I think that makes sense.
73.
by rileyphone 3y ago
You should check out uvu
74.
by rileyphone 3y ago
It really depends on your approach to learning. If you're just learning in the rubric style, taking a checklist of topics, assignments, like school, yeah that will reduce your creativity. But if you're just exploring and playing w
75.
by rileyphone 3y ago
the original wiki
76.
by rileyphone 3y ago
You can run it (quantified at least) on a $4000 Mac thanks to Apple's unified memory. Surely other manufacturers are looking for how to expand VRAM, hopefully Intel or AMD.
77.
by rileyphone 3y ago
Maybe Austral? https://austral-lang.org/
78.
by rileyphone 3y ago
In what way is Smalltalk similar? Its main problem was that it stopped evolving.
79.
by rileyphone 3y ago
The vast majority of humans would fail your reasoning test.
80.
by rileyphone 4y ago
I’m working on a syntax for an object oriented lisp-like thats basically this (the self receiver is implied) .myfunction(.a .b.+(.c).+(.d).+(.f) .g) though making binary arithmetic operations singly-dispatched methods of the number
81.
by rileyphone 4y ago
Basically fexprs
82.
by rileyphone 4y ago
You can use verified domains as handles
83.
by rileyphone 4y ago
This happened with ChatGPT a few weeks ago. https://news.ycombinator.com/item?id=35291112
84.
by rileyphone 4y ago
Isn't available in Firefox though...
85.
by rileyphone 4y ago
It took him a while to come around https://en.wikipedia.org/wiki/An_Open_Letter_to_Hobbyists
86.
by rileyphone 4y ago
> it isn't smart enough to thoughtfully make comments about how data is pipelined through a complete application My take is that data-centric programming requires too much context for GPT, and we're going to see a move back to
87.
by rileyphone 4y ago
USD
88.
by rileyphone 4y ago
it’s like “world peace”
89.
by rileyphone 4y ago
It worked for Instagram - Clubhouse's failure was far more spectacular.
90.
by rileyphone 4y ago
Would recommend https://mynoise.net/noiseMachines.php specifically, they have a bunch of great sounds.

More ›