Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
BadBadJellyBean
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
31.
▲
by
BadBadJellyBean
3mo ago
The problem I see with the change in their rules is that they have created an unenforceable or worse arbitrarily enforced rule. There is no sure way to identify LLM code. I could write a project by hand with all the GPT "mannerisms&quo
32.
▲
by
BadBadJellyBean
3mo ago
> The thing is not that the product will be somehow secure. I don't believe that is true. Cars do break but they have to meet safety standards. That doesn't make them perfect but it's much better than not having them. You
33.
▲
by
BadBadJellyBean
3mo ago
Just like everyone left when the GDPR went into place? Having safety standards in place (and yes that also means documenting them) can even be a selling point. To have demonstrably secure hardware according to a standard can be highly desi
34.
▲
by
BadBadJellyBean
3mo ago
Also apps need to be allowed to conduct their transactions outside the appstore and the appstore must allow them to advertise that.
35.
▲
by
BadBadJellyBean
3mo ago
I'd rather have that than insecure hardware. It seems like it forces security to be a first class priority instead of an afterthought.
36.
▲
by
BadBadJellyBean
3mo ago
Because they decided that way. Any other country in the world can decide the same thing but they don't have the same leverage as the EU. Google makes a lot of money in the EU.
37.
▲
by
BadBadJellyBean
3mo ago
The maximum fine the EU can impose with the DMA is "10 % of its total worldwide turnover in the preceding financial year". I think this would hurt even Google. And I think a billion dollars is not nothing.
38.
▲
by
BadBadJellyBean
3mo ago
I'm not familiar with the Act. Can you explain how it would kill theses startups?
39.
▲
by
BadBadJellyBean
3mo ago
They are bigger. Not as easy to guess. More like pretty impossible. It's like not letting the user choose the password. That way they can't have a bad password.
40.
▲
by
BadBadJellyBean
3mo ago
Nope. I started with old school blender.
41.
▲
by
BadBadJellyBean
3mo ago
I think you need to start learning a lot of things from scratch. They are just very different programs with different concepts and layouts. You can sure transfer some things over but I think it might be best to just start from scratch, skip
42.
▲
by
BadBadJellyBean
3mo ago
I started with blender and then tried C4D. Never got the hang of it. Familiarity is really a strong factor.
43.
▲
by
BadBadJellyBean
3mo ago
This! Who types full commands?
44.
▲
by
BadBadJellyBean
3mo ago
err := do_something() if (err != nil) { return nil, err } err = do_something_else() if (err != nil) { return nil, err } err = do_something_more() if (err != nil) { return nil, err } ... So very concise.
45.
▲
by
BadBadJellyBean
3mo ago
There is more to it. To request information you have to have some "legitimate interest" which is not yet defined and sounds like an easy way to not give out information. Also no more NGOs. They were the ones who had the most power
46.
▲
by
BadBadJellyBean
3mo ago
Declaration of intent is a good description. Nothing is decided yet. I don't think the laws are written yet and there was definitely no vote yet.
47.
▲
by
BadBadJellyBean
3mo ago
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.
48.
▲
by
BadBadJellyBean
3mo ago
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.
49.
▲
by
BadBadJellyBean
3mo ago
But LLMs can do it much faster and more consistently.
50.
▲
by
BadBadJellyBean
3mo ago
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 implementatio
51.
▲
by
BadBadJellyBean
3mo ago
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 se
52.
▲
by
BadBadJellyBean
3mo ago
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 fin
53.
▲
by
BadBadJellyBean
3mo ago
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 mi
54.
▲
by
BadBadJellyBean
3mo ago
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 b
55.
▲
by
BadBadJellyBean
3mo ago
CoD is by Activision. They are owned my Microsoft.
56.
▲
by
BadBadJellyBean
3mo ago
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 mea
57.
▲
by
BadBadJellyBean
3mo ago
Why not both?
58.
▲
by
BadBadJellyBean
4mo ago
Airports could potentially be such a place but I admit that this is a bit contrived.
59.
▲
by
BadBadJellyBean
4mo ago
Well I didn't see any counter evidence by Planck so Springer Nature must be right. Otherwise he should defend himself! \s
60.
▲
by
BadBadJellyBean
4mo ago
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.
More ›