danieltanfh95

6 ms·

61.
by danieltanfh95 6mo ago
I want to make my stance clear. Bell's theorem rules out local hidden variables, not hidden variables. An ontology-first approach to QM (rare in modern physics) takes Bell as the constraint it has to operate inside (same as Bohm-de Bro
62.
The Century-Long Pause in Fundamental Physics(danieltan.weblog.lol)
27 points by danieltanfh95 6mo ago|19 comments
63.
by danieltanfh95 6mo ago
Sigh. I meant that the zig authors did not make it a general pattern and just slapped on the DI pattern specifically for io, instead of generalising the abstraction so people can DI stuff.
64.
by danieltanfh95 6mo ago
A common workflow is to run code to test some function in the REPL and then promote it to a test when you are ready, and this process has been the smoothest in lisps, especially since you can create your own test harness if you need to. Fun
65.
by danieltanfh95 6mo ago
I am not even sure if its a general pattern (inject any dependency?) or a specific pattern they added to Zig
66.
by danieltanfh95 6mo ago
Which is why Clojure advocates for simple data structures. They compose easier than code itself.
67.
by danieltanfh95 6mo ago
That's what I started doing in my company, encouraging the use of AI to polish prototypes and communication before handing it to the tech team smooths out alot of issues while exloring the solution space.
68.
Rules, Not Weights(danieltan.weblog.lol)
1 points by danieltanfh95 6mo ago|0 comments
69.
by danieltanfh95 6mo ago
https://github.com/danieltanfh95/replsh LLMs are surprisingly bad at using REPLs, so I made a CLI that handles sync, streaming, async REPL evals support over docker, ssh, local, and supporting python and clojure. Also
70.
by danieltanfh95 6mo ago
clojure exists as an example of people trying types and then realising it's cruft and not needed.
71.
by danieltanfh95 6mo ago
No, it is evidence for that point. You could just rattle off every possible vulnerability and have the cheap model scan for it in the harness through a loop. Note that I say cheap, not small, because small models may lack the reasoning need
72.
by danieltanfh95 6mo ago
People really lack imagination. The point here is that a dedicated attacker with a good harness and really cheap models can run the attack regardless. It's like portscan/url search attacks. They could run all of these against all
73.
The Case That More Openness Brings More Good to Society(danieltan.weblog.lol)
2 points by danieltanfh95 6mo ago|1 comments
74.
by danieltanfh95 6mo ago
interesting, i also went the same direction with https://github.com/danieltanfh95/replsh because i needed to support existing legacy python codebases.
75.
by danieltanfh95 6mo ago
built https://github.com/danieltanfh95/replsh to pair with local python sessions without additional dependencies, allowing LLMs to directly ground their investigation and coding against local repos and environments. No
76.
by danieltanfh95 6mo ago
I think the discussion has to be more nuanced than this. "LLMs still can't do X so it's an idiot" is a bad line of thought. LLMs with harnesses are clearly capable of engaging with logical problems that only need text. L
77.
Give LLMs a Thinking Medium(github.com)
1 points by danieltanfh95 6mo ago|0 comments
78.
Succession – Agent Lineage Evolution for an Agentic World(danieltan.weblog.lol)
1 points by danieltanfh95 7mo ago|0 comments
79.
by danieltanfh95 7mo ago
How is error handling expected to happen here?
80.
The Armor Gap(danieltan.weblog.lol)
1 points by danieltanfh95 7mo ago|0 comments
81.
Pokemon and the Iran War Nerdsniped Me into Quantifying Strategy(danieltan.weblog.lol)
2 points by danieltanfh95 7mo ago|0 comments
82.
by danieltanfh95 7mo ago
Maybe the more rational conclusion is that it is purely a random chance that a startup would succeed, so we should just increase the amount of startups in the first place instead of restricting it.
83.
by danieltanfh95 7mo ago
Really the same mechanics with crypto
84.
by danieltanfh95 7mo ago
Agentic engineering is working from documentation -> code and automating the translation process via agents. This is distinct from the waterfall process which describes the program, but not the code itself, and waterfall documentation ca
85.
by danieltanfh95 8mo ago
It is just search. If it can't even execute search then what else would it be good at?
86.
by danieltanfh95 8mo ago
Hallucinates like crazy. use with caution. Tested it with a simple "Find me championship decks for X pokemon", "How does Y deck work". Opus 4.6, Deepseek and Kimi all performed well as expected.
87.
by danieltanfh95 8mo ago
> static types are not about “classifying the world” or pinning down the structure of every value in a system. The reality is that static type systems allow specifying exactly how much a component needs to know about the structure of its
88.
by danieltanfh95 8mo ago
python, js, ruby, clojure, erlang etc are created by senior engineers who are fully aware that haskell etc exists, so have you considered that people you claim are "senior" aren't senior enough?
89.
by danieltanfh95 8mo ago
People keep saying this and yet in the decades of my career the industry bounces between being fully dynamic and fully typed according to the affordability of senior engineers. What you are saying are covered by tests, not types.
90.
by danieltanfh95 8mo ago
"good type systems" in Python and Javascript has a escape hatch via dicts/objects, and are frankly, not "good" compared to FP via haskell etc, but it is realistic, reliable, open and consistent, which is exactly my

More ›