erichocean

63 ms·

61.
by erichocean 4mo ago
> Half the code was written in a language you didn't understand. The other half was written using libraries you never heard of. The author already describes himself as "not a rockstar developer", but if this is the defin
62.
by erichocean 4mo ago
I wish you guys could partner with Modular to get Mojo inference working on your hardware, e.g. https://www.modular.com/models/deepseek-v4-pro
63.
by erichocean 5mo ago
There are deep reasons for the variations, especially around (reader) macros.
64.
by erichocean 5mo ago
I suspect the author created his own IR after being offered that suggestion earlier, he's definitely aware it exists.
65.
by erichocean 5mo ago
People also liked Arcane.
66.
by erichocean 6mo ago
I wish they offered a Dvorak keyboard. Of all laptops, this is the most obvious one to do it.
67.
by erichocean 6mo ago
Humans have, obviously. Just interbreeding with ancient species was enough to do it, even without separate evolution.
68.
by erichocean 6mo ago
Autochrome [0] make reviewing Clojure diffs from AI really easy. [0] https://github.com/ladderlife/autochrome
69.
by erichocean 6mo ago
I also interact a ton with C and C++, but it's easy today to have Claude write a Project Panama wrapper and then put a nice Clojure veneer on top of the Java.
70.
by erichocean 6mo ago
> it just couldnt write the changes to file 9/10 times when using it with opencode You might want to give this a try, it dramatically improves Edit tool accuracy without changing the model: https://blog.can.ac/2026
71.
by erichocean 6mo ago
Another limitation: only five active user accounts (with UIs) per machine.
72.
by erichocean 6mo ago
> Is that what you mean? I think the confusion is that, when I write "model", you read "LLM." LLMs aren't the only kind of AI model, and they have the limitations Aphyr mentions, for the obvious reasons you&#x2
73.
by erichocean 6mo ago
> Models do not (broadly speaking) learn over time. They can be tuned by their operators, or periodically rebuilt with new inputs or feedback from users and experts. Models also do not remember things intrinsically: when a chatbot refer
74.
by erichocean 6mo ago
AI is exactly the right term: the machines can do "intelligence", and they do so artificially. Just like we have machines that can do "math", and they do so artificially. Or "logic", and they do so artificially
75.
by erichocean 6mo ago
As I remember it, they were basically the same—but IOKit is C++ (with restrictions) because 3rd party developers didn't want to learn Objective-C. But that's a hazy, 20 year old memory.
76.
by erichocean 6mo ago
> Isn't it more-or-less the same as the Python REPL? Not even close.
77.
by erichocean 6mo ago
Also discussed in detail in Crafting Interpreters.[0] [0] https://craftinginterpreters.com/
78.
by erichocean 6mo ago
> Bridging that mismatch at the macro level seems like the harder problem than the basic REPL integration. You can (and people do, core.async in Clojure works this way) put entire compilers in macros, macros are just functions that tak
79.
by erichocean 7mo ago
How about this? (filter ds (> year 2008)) That's a trivial Clojure macro to make work if it's what you find "intuitive."
80.
by erichocean 7mo ago
> is just so much more readable I thought that too before I learned Clojure, now I find them equally readable.
81.
by erichocean 7mo ago
What I'd like to see is Omarchy implemented via the Nix package manager. (Seems like a good project for AI, actually.)
82.
by erichocean 7mo ago
> If you know of any other snippet of code that can master all that complexity as beautifully, I'd love to see it. Electric Clojure: https://electric.hyperfiddle.net/fiddle/electric-tutorial.tw...
83.
by erichocean 7mo ago
They don't read the things you send them, not even for "safety checks" or sys-admins accessing the system. Totally opaque (as it should be). Keeping chat content around for 30 days might as well mean "forever." Anyo
84.
by erichocean 7mo ago
> what would make this actually useful for you? A privacy policy that's at least as good as Vertex.ai at Google. Otherwise it's a non-starter at any price.
85.
by erichocean 7mo ago
Ideal HN content, thanks!
86.
by erichocean 7mo ago
Partially true, you can predict multiple tokens and confirm, which typically gives a 2-3x speedup in practice. (Confirmation is faster than prediction.) Many models architectures are specifically designed to make this efficient. --- Separat
87.
by erichocean 7mo ago
> "something about which no conclusions can be drawn because the proposed definitions lack sufficient precision and completeness." The same problem exists defining human intelligence, it's a problem with "intelligen
88.
by erichocean 7mo ago
Saying CGI is animation is a bit like saying a photo printer prints paintings, because the latter uses pigments (just like painters!).
89.
by erichocean 7mo ago
Yup, I had Claude write a tool to auto-fix those diagrams. :D
90.
by erichocean 8mo ago
I would give it every example TLA+ file you can find, including PlusCal stuff (as reference). Also provide an up to date language manual and/or grammar if you have one.

More ›