zahlman

40 ms·

811.
by zahlman 4mo ago
It's the simplest model that gives the right result for simple cases (e.g. once one pile is empty, the remaining cards must come from the other pile; and when they're evenly split, it should be a coin flip). It also entails, for e
812.
by zahlman 4mo ago
It requires eight perfect riffle shuffles, not seven. (I just checked at the Python REPL.) And actually it depends on whether the riffles are done "in" or "out" (i.e. which half of the deck the new top card comes from).
813.
by zahlman 4mo ago
I have a new-repository script that creates a .local directory and puts a .gitignore with just `*` in it.
814.
by zahlman 4mo ago
FWIW I find that GPT can be very creative when discussing a high-level design. Once it starts writing code snippets it will offer to take things in a bunch of different directions.
815.
by zahlman 4mo ago
> being nice to Claude will be rewarded and being mean to Claude will be punished ... That does sound like something that Anthropic would deliberately aim for, yeah. > With GPT, you have to be precise and reduce ambiguity. I have foun
816.
by zahlman 5mo ago
> No X. No Y. Just Z. is another noted LLM-ism.
817.
by zahlman 5mo ago
> even an ipad with a whiteboard app aren't good replacements for me. Curious. My first thought would have been "... because the line is physically displaced from the pen, of course"; but that doesn't apply there.
818.
by zahlman 5mo ago
I think agreement has value here too. An LLM that's starting to get a bit sycophantic will rephrase your ideas in a few different ways, and seeing the different presentations is helpful for reconsideration.
819.
by zahlman 5mo ago
From my own experience chatting with LLMs: reading the responses definitely does help with thinking, even when you can see obvious flaws or hallucinations. It gives you something to think about , which a rubber duck can't do.
820.
by zahlman 5mo ago
Eh, the results for Julia seem impressive enough.
821.
by zahlman 5mo ago
Why shouldn't the author mention models that people might not have to buy a new computer to use?
822.
by zahlman 5mo ago
> The hilarious part, though, is that it's not the AI that's working around the rules. That's the scenario that's been in science fiction, but it's not what's happening. It's the human users making use
823.
by zahlman 5mo ago
Many, many years ago I was asked to implement a filter like that for usernames. I said right away that it wasn't going to work well, but I did implement it. Next internal build, the CEO can't create an account. With his real name.
824.
by zahlman 5mo ago
I think I'm not getting something here. Like, sure, the refused prompt "review the code for security issues" could be interpreted as an attempt to discover weaknesses in a running system to exploit them. But we don't ge
825.
by zahlman 5mo ago
> you have to "crank" your scroll wheel to get to the content, just like you have to crank this device. I actually didn't consider that interpretation. (I also almost deleted the comment, definitely didn't expect it t
826.
by zahlman 5mo ago
The webpage linked is an example of everything I wish people would stop doing in web design. Fortunately, at the bottom there is a link to the "technical documentation" ( https://squeezlabs.github.io/handcrank/
827.
by zahlman 5mo ago
> The swing of the market based on the president's crazy tweets is just insane. A quick glance at charts shows that VIX is not at all out of line with historical patterns, and asking ChatGPT to crunch some numbers confirms that. The
828.
by zahlman 5mo ago
> Otherwise there is so, so much evidence that it is flat wrong. Pardon; are you asserting there is mountains of evidence that the bears have not, in fact, claimed impending recession far more often than actual recession occurred? Or tha
829.
by zahlman 5mo ago
It took me quite a while to figure out that the two of you are using "port" as short for "portfolio". Never heard that before. "The wealthy people" got wealthy in a whole bunch of different ways that are not in
830.
by zahlman 5mo ago
> 99. In man-machine symbiosis, it is man who must adjust: The machines can't. It seems worth noting here that the English verb "to adjust" is ambitransitive.
831.
by zahlman 5mo ago
But don't you still have to logically connect the validity of the proof to desirability of the output?
832.
by zahlman 5mo ago
> how effective frontier models (ChatGPT-5.5 in particular) are at completing certain manual proofs in the Roqc (né Coq) proof assistant. The proofs aren't always pretty, but ChatGPT can often prove something in minutes and 10 - 100
833.
by zahlman 5mo ago
> But the ridiculousness of saying New York style pizza is not pizza or that you have to make things the "right way" needs to go . Suppose people say it; why shouldn't they be entitled to their opinion? How does it harm
834.
by zahlman 5mo ago
> or example you had early Scottish shortbread primarily using rice flour because it was cheaper at that time. What era of history are we talking about here? Would it have been transported as flour, or ground locally?
835.
by zahlman 5mo ago
Time in the market beats timing the market, bears have predicted (large number) out of the last (smaller number) recessions, etc. None of this is novel. "Predicting" a stock crash due to political reasons is effectively just a fan
836.
by zahlman 5mo ago
This is not any better than the reply about Obama. The intention of the snark is clear, and does not contribute to useful discourse. Please don't.
837.
by zahlman 5mo ago
> Please don't use Hacker News for political or ideological battle. It tramples curiosity.
838.
by zahlman 5mo ago
Institutional investors get to trade outside of hours that most ordinary people can't; presuming that this is net positive for markets, they will already be up when the bell rings.
839.
by zahlman 5mo ago
Markets reached ATH this month and are only a couple percent off that mark. What is there to "quell"?
840.
by zahlman 5mo ago
I love TUIs. Anthropic failing at making them properly is not the fault of the paradigm.

More ›