maxbond

22 ms·

361.
by maxbond 11mo ago
I don't think there was very much abuse of "not just A, but B" before ChatGPT. I think that's more of a product of RLHF than the initial training. Very few people wrote with the incredibly overwrought and flowery style o
362.
by maxbond 11mo ago
I'm also concerned whether it is useful! That's why I'm not gunnuh read it after receiving a strong contrary indicator (which was less the use of AI than the dishonesty around it ). That's also why I try to avoid soundi
363.
by maxbond 11mo ago
I understand being okay with a book being generated (some of the text I published in this manual [1] is generated), I can imagine not caring that the author lied about their use of AI, but I really don't understand the suggestion I wri
364.
by maxbond 11mo ago
So petty as to lie about using AI or so petty as to call it out? Calling it out doesn't seem petty to me. I intend to learn Zig when it reaches 1.0 so I was interested in this book. Now that I see it was probably generated by someone w
365.
by maxbond 11mo ago
You can be skeptical of anything but I think it's silly to say that these "Not just A, but B" constructions don't strongly suggest that it's generated text. As to why it matters, doesn't it matter when people l
366.
by maxbond 11mo ago
Using AI to write is one thing, claiming you didn't when you did should be objectionable to everyone.
367.
by maxbond 11mo ago
I think the problem with these examples is that they conflate instrumental goals with terminal goals. People speed to get to a destination faster or to relieve their frustration on the road (street racers notwithstanding). If the cost of sp
368.
by maxbond 11mo ago
My bad for misreading you
369.
by maxbond 1y ago
https://peps.python.org/pep-0810/#what-about-star-imports-fr... > What about star imports (`from module import *`)? > Wild card (star) imports cannot be lazy - they remain eager. This is because the set of names
370.
by maxbond 1y ago
If you succeed in creating a generic async primitive, it doesn't really matter what the original task was (as long as it's something that requires async), no? That's an implication of it being generic?
371.
by maxbond 1y ago
Prompt engineering ages like milk and doesn't hold a child's attention. The PB&J demo is evergreen and hilarious.
372.
by maxbond 1y ago
For this type of thing I think property testing would work well. It would take a fair about of work to write a proptest for the entire input space of the tool. But it's achievable and as durable as the CLI arguments (so for this specif
373.
by maxbond 1y ago
> I guess there are people who plan everything in their mind and the coding part is just typing out their ideas (instead of developing their ideas during code editing). I don't think there are, I think Gall's law that all compl
374.
by maxbond 1y ago
People Make Games did a story on this market. https://www.youtube.com/watch?v=eMmNy11Mn7g [36m]
375.
by maxbond 1y ago
A good compromise would be only marking their top-level comments. And/or doing this only for Show/Ask/Tell threads. I'm sending this thread to the admins as a feature request.
376.
by maxbond 1y ago
There are some vulnerabilities frameworks can address wholesale (like CSRF or XSS) as long as you keep to the blessed way of doing things, but they aren't able to save you from a complete failure to build authorization into your API. L
377.
by maxbond 1y ago
`bcrypt` is probably the "standard" in the sense that it has the widest adoption, but since 2015 [1] the "standard" in terms of what you should recommend for new work has been `argon2id` (and you can find parameter recom
378.
by maxbond 1y ago
> [H]ow do I know the thing you checked or the context in which it was checked is any better than my doing nothing? Open the paper and look for it's methodology.
379.
by maxbond 1y ago
This is a bit of a "no true Scotsman" issue. Almost no one working as a journalist is given the resources to do that. Even if they have access to those resources, they don't necessarily have access to them for every story .
380.
by maxbond 1y ago
> ...but for some reason journalists don't do this. I imagine most journalists would love to have technical reviewers on their work, but there's no funding for it and there's pressure to churn content as quickly as possibl
381.
by maxbond 1y ago
For macros, yes, `cargo expand`. For generics, no, but you can query this information in your IDE pretty easily. You can also also use `cargo doc` to build the generated documentation, which will show you the list of traits something implem
382.
by maxbond 1y ago
It's called PROJECT INTO in this language.
383.
by maxbond 1y ago
I think they were referring to intermediate tokens in "Thinking" models, which are summarized in the interface but ultimately discarded (and may themselves be summaries of sources, other chats, or earlier intermediate states). Pre
384.
by maxbond 1y ago
What do you think would be a better estimate for average salary if we assume heavy offshoring?
385.
by maxbond 1y ago
Even if you are profitable your investors could pressure you to exit to some other company so that they can recoup their investment. But I doubt they were profitable. Going by the article they had about 100 employees and about $2M in revenu
386.
by maxbond 1y ago
> I guess we can characterize this as some kind of hallucination+streisand effect combo... I would call it citogenesis or circular reporting . Or perhaps machine citogenesis or model citogenesis . https://xkcd.com/9
387.
by maxbond 1y ago
I've vouched for your comment. I don't see any reason why your comment would require vouching, so I would guess that some overzealous spam filter flagged it. I would recommend emailing hn@ycombinator.com to ask about the status of
388.
by maxbond 1y ago
Petition the Unicode consortium to include a seahorse emoji.
389.
by maxbond 1y ago
As in microcontrollers, it's a silly pun.
390.
by maxbond 1y ago
You know what they call someone who studied computer science and was a C student? An embedded engineer.

More ›