Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
zer00eyz
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
27 ms
·
481.
▲
by
zer00eyz
1y ago
> I’ve never seen such glue code that is tedious to write. Its a fair point, its not the writing per se thats tedious: Fetch data from API 9522, write storage/trasformation/validation code, write display code. Test, tweak/
482.
▲
by
zer00eyz
1y ago
The pro-LLM crowd on HN is just as cultish. The divide is as diverse as the work we do: There is work that I do that is creative, dynamic and "new". The LLM isn't very helpful at doing that work. In fact it's pretty bad
483.
▲
by
zer00eyz
1y ago
Chiron: 2004, the UK government shut down their flu vax plant (it was in the UK). It later came out that the FDA knew what was up and basically let it slide. It was one of the early ani-vax movements torches... Crunchy moms pissed about sho
484.
▲
by
zer00eyz
1y ago
> this feels like a fundamental misunderstanding of how video game dialogue writing works. Factorio, Dwarf Fortress, Minecraft. There are plenty of games where the whole story is driven by cut scenes. There are plenty of games that shove
485.
▲
by
zer00eyz
1y ago
> If the top 1% of companies drive the majority of VC returns The fact that the author brings this up and fails to realize that the behavior of current staff shows we have hit or have passed peak AI. Moores Law is dead and it isn't
486.
▲
by
zer00eyz
1y ago
So because it has a foreign name its not rich influencer nonsense? > The estimated cost for a pui yuet in Hong Kong ranges from 63,800 Hong Kong dollars (US $8,100) to 268,000 Hong Kong dollars (US $34,100) for 26 to 30 nights for a live
487.
▲
by
zer00eyz
1y ago
> And I'm guessing just based on my own experience paying for term life that the actual premia differences aren't actually enough in most cases for the life insurer to simply pay out of pocket themselves; the differences probab
488.
▲
by
zer00eyz
1y ago
> If you’re working with goroutines, you would always pass in a context parameter to handle cancellation. The utility of context could be called a subtle coloring. But you do NOT need context at all. If your dealing with data+state (arou
489.
▲
by
zer00eyz
1y ago
> We need to make it a lot cooler for researchers to not do original research all the time but to do replication and peer review. This should be the work of grad students, not cranking out another paper or slaving for professors. > We
490.
▲
by
zer00eyz
1y ago
I don't think in this case your argument is as clear cut and the use cases that people have today arent solved by the choices out there. George Carlin: "You don't need a formal conspiracy when interests converge. These people
491.
▲
by
zer00eyz
1y ago
> I get the desire to reduce government spending. It looks like sticker shock seeing budgets in the billions and trillions. This is an international issue being funded by the US taxpayer regardless of their own utilization of said servic
492.
▲
by
zer00eyz
1y ago
> The browser’s access to a user’s web activity would make it the ideal platform for AI “agents” that can take actions on their behalf, like booking reservations or filling out forms, directly within the websites they use. How about agen
493.
▲
by
zer00eyz
1y ago
>1) we will live with the reduced usefulness of models trained on data that has been poisoned This is the entirety of human history, humans create this data, we sink ourselves into it. It's wishful thinking that it would change. >
494.
▲
by
zer00eyz
1y ago
> Now, why are you spending misinformation? Are you going to claim that US politicians don't do the exact same thing? This is my favorite example of it, where one literally tells you what the play is while it's getting made:
495.
▲
by
zer00eyz
1y ago
> including systems that can evaluate news sources, understand satire, and so forth. Lets take something that has been in the news recently: https://abcnews.go.com/Business/wireStory/investors-snap-gro... "
496.
▲
by
zer00eyz
1y ago
> if AI capabilities taper off AI growth has slowed to a crawl, and it's priced it self out vs cost of compute. NVIDIA feels a lot like SUN. > amazing documentary Been there, done that: 2001, Startup Dot Com https://ww
497.
▲
by
zer00eyz
1y ago
Cantillon's essay is not terribly difficult of a read and the "Cantillon Effect" has to be the least interesting part of it. It and Smiths Wealth of Nations are both free on the web and well worth the read.
498.
▲
by
zer00eyz
1y ago
I can't remember the last time I went out of my way to look at who authored something because I disliked it on a visceral level. The writing here goes from too much punctuation to grad students book review to quasi political rant. And
499.
▲
by
zer00eyz
1y ago
https://archive.is/Py9Mf
500.
▲
by
zer00eyz
1y ago
> There is obviously a difference between understanding what smarter people have already done, and actually innovating and building on what they have done. The latter requires a much deeper understanding. Does it really? When you build a
501.
▲
by
zer00eyz
1y ago
https://www.wired.com/story/ai-art-copyright-matthew-allen/ https://www.cnbc.com/2025/03/19/ai-art-cannot-be-copyrighted... Here are cases where the product of AI/ML are not th
502.
▲
by
zer00eyz
1y ago
> or public domain https://news.artnet.com/art-world/ai-art-us-copyright-office... https://en.wikipedia.org/wiki/Monkey_selfie_copyright_disput... Im pretty sure that this ship has sailed.
503.
▲
by
zer00eyz
1y ago
This is a spurious argument at best: There isnt a line of unemployed draftsmen out there begging for change cause we invented Autocad: https://azon.com/2023/02/16/rare-historical-photos/ What happened to
504.
▲
by
zer00eyz
1y ago
> and a pretty short view of history Great lets see an example! > To claim that all of the benefit isn't going to naturally accrue to a thin layer of people at the top isn't a speculation at this point - it's a bold fac
505.
▲
by
zer00eyz
1y ago
No, they do not. No LLM is ever going to be self aware. It's a system that is trained, that only does what you build into. If you run an LLM for 10 years it's not going to "learn" anything new. The whole industry needs t
506.
▲
by
zer00eyz
1y ago
> seems more philosophical than scientific I think this is a fair assessment but reason, and intelligence dont really have an established control or control group. If you build a test and say "Its not intelligent because it can'
507.
▲
by
zer00eyz
1y ago
> drug LSD, which is used to treat conditions like depression, post-traumatic stress disorder and addiction. Umm, ok, sure thats what it's used for. That one sentence just points to a large issue with the whole tone of this article.
508.
▲
by
zer00eyz
1y ago
I see someone has heard from their legal department on AI tools.
509.
▲
by
zer00eyz
1y ago
Because much of what passes as "security" is a bunch of theater. > SMS gets forwarded to e-mail, TOTP codes get sent over Wechat, Here we are deep into 2FA land. Where you have institutions blocking SMS/MMS to IP telephony
510.
▲
by
zer00eyz
1y ago
> It also begs the question of whether there exists a clear and narrow definition of what "thinking" is that everyone can agree on. I suspect if you ask five philosophers you'll get six different answers, as the saying goe
More ›