HackerTrans
TopNewTrendsCommentsPastAskShowJobs

BadBadJellyBean

1,027 karmajoined 7 years ago
[ my public key: https://keybase.io/cperrin; my proof: https://keybase.io/cperrin/sigs/rDhx9kd_MoMMLj8HV0TlgVdk4YUQGgp4vctPXINy51g ]

Submissions

[untitled]

1 points·by BadBadJellyBean·26 days ago·0 comments

comments

BadBadJellyBean
·yesterday·discuss
This looks great. We have a lot of bash tools because it's the only stable interpreter that we have readily available on all our systems. But bash is a pain to write so this might actually make things easier.
BadBadJellyBean
·2 days ago·discuss
Since the whole leap second system will be phased out by 2035 anyways I doubt that anyone will test it. No need to rock the boat over a second.
BadBadJellyBean
·2 days ago·discuss
But LLMs can do it much faster and more consistently.
BadBadJellyBean
·3 days ago·discuss
One problem there is that even if the code is consistent, idiomatic Rust is not the same as idiomatic C.

So either the translation produced Rust code in the "shape" of C code or the code is quite different from the C implementation.

Add to that that you need to have people proficient in Rust and also Postgres as well as the very much unknown codebase as a whole you get a recipe for pain.
BadBadJellyBean
·3 days ago·discuss
I don't think you gain anything. The article is written, it will not be rewritten or unwritten because you said it's written by an LLM. I don't see the author not using LLMs in the future if they did use them.

I don't see how you changed anything by pointing out that you assume that this is written by an LLM. What was gained? I don't really see anything. Do you think anyone is dissuaded from reading the article or from using an LLM the next they write something because you could maybe probably tell that it's probably written by an LLM?

I think we all lose more than we gain if we do this. We look at each other with mistrust over whether they used an LLM and we point fingers as soon as we see a sign of "wrongdoing". I see artists and writers frustrated about being accused of using ML tools. And the people who just use LLMs and image generation tools mostly just don't care.

The whole "AI" industry makes everything diffuse. We have to go by vibes if something is made by a human or not and the signs are ever more subtle. We have to be so incredibly careful not to accuse those who are on our side. People who create real human work.

For me the easiest solution for now is to stop accusing. Unless you know for certain that there was not a real human behind it don't point fingers.
BadBadJellyBean
·3 days ago·discuss
You can't reliably prove that something is written by an LLM. There are certainly tells but it could be a personal writing style as well. When reading everything with the suspicion that it might be written by an LLM you are at best finding LLM written content and at worst accuse people of using an LLM when they haven't. Nothing is gained by the accusation.

For me a better way is to find out if I want to read the text or not. Does is there something interesting being said? Is it presented in a way that is at least pleasant enough to read? Is it concise enough? If not I don't read it. Or I skim it.

Don't misunderstand me. I don't like the overuse LLM generated texts. I write my words on my own. Still there is nothing to be gained from distrusting and calling out authors.
BadBadJellyBean
·3 days ago·discuss
Maybe it is. Maybe it isn't. You gain nothing from pointing out every post that seems LLM generated. Read it or don't but we don't make the world better by accusing each other of using LLMs. The only things we increase are mistrust and frustration.
BadBadJellyBean
·4 days ago·discuss
It's not enough to do a rewrite. Someone has to maintain it. Such a huge codebase with literally zero experts is unmaintainable. There is no one who knows how the internals work.

Sure you could keep vibe coding it but I wouldn't bet my data on that. A database needs to be rock solid.
BadBadJellyBean
·6 days ago·discuss
CoD is by Activision. They are owned my Microsoft.
BadBadJellyBean
·6 days ago·discuss
I was always very skeptical of the $300 million figure. It sounded like the same math that the movie industry used for pirating. If I was subscribed to Game Pass I might have downloaded CoD to see what it's about. That doesn't mean I would have paid full price for it.
BadBadJellyBean
·11 days ago·discuss
Airports could potentially be such a place but I admit that this is a bit contrived.
BadBadJellyBean
·16 days ago·discuss
Well I didn't see any counter evidence by Planck so Springer Nature must be right. Otherwise he should defend himself! \s
BadBadJellyBean
·18 days ago·discuss
PostgreSQL is a powerhouse. It has a solution for everything. Especially when you start a project you might be better off just using PostgreSQL instead of a specialized solution. You can optimize it later.
BadBadJellyBean
·18 days ago·discuss
It's for those people that can't wait a second after release. As if the game is any worse if you start playing it a few days later.
BadBadJellyBean
·19 days ago·discuss
I'm rooting for the EU.
BadBadJellyBean
·19 days ago·discuss
VISA has debit cards as well. I have that.
BadBadJellyBean
·19 days ago·discuss
I don't know why but my bank offers a VISA debit card and an EC card. I can use the VISA card on every ATM without fees and it doesn't cost me extra. The EC card has extra fees for when I want to get cash with it. I have seen other banks do the same. I think I heard that this is a VISA thing but maybe I'm wrong or misheard.
BadBadJellyBean
·19 days ago·discuss
True but that doesn't seem to be a thing here.
BadBadJellyBean
·19 days ago·discuss
Germany is different. I don't need cash often but sometimes it's needed.
BadBadJellyBean
·19 days ago·discuss
I don't know how but AFAIK VISA has a contract clause that prohibits the issuing bank from charging their customers fees for using other ATMs.