Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
zmmmmm
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
30 ms
·
541.
▲
by
zmmmmm
1y ago
Which means they had at best shallow involvement and left the scene pretty quickly. Probably no realistic idea whether the systems created survived long term impact with reality or not. But hey, free advice!
542.
▲
by
zmmmmm
1y ago
It says something that he still believes he has "missionaries" after betraying all the core principles that OpenAI was founded on. What exactly is their mission now other than generating big $?
543.
▲
by
zmmmmm
1y ago
It a seems pretty simple mistake from almost the first step: > To get funding, you need users. To get more funding, you need exponential growth. Growth becomes the primary metric No, you don't. This exponential user scaling theory o
544.
▲
by
zmmmmm
1y ago
I think there's a key context difference here in play which is that AI tools aren't better than an expert on the language and code base that is being written. But the problem is that most software isn't written by such expert
545.
▲
by
zmmmmm
1y ago
yeah ... it's going to depend how the issue is framed. However a "copy" of something where there is no way to practically extract the original from it probably has a pretty good argument that it's not really a "copy
546.
▲
by
zmmmmm
1y ago
It's important to note the way it was measured: > the paper estimates that Llama 3.1 70B has memorized 42 percent of the first Harry Potter book well enough to reproduce 50-token excerpts at least half the time As I understand it, i
547.
▲
by
zmmmmm
1y ago
interesting choice - doesn't that then mean that container to container integration is going to be harder and a lot of overhead per-container? I would have thought a shared VM made more sense. I wonder what attracted them to this.
548.
▲
by
zmmmmm
1y ago
Gemma 12b quantized (gemma3:12b-it-qat in ollama)
549.
▲
by
zmmmmm
1y ago
I think in a lot of cases the skeptics haven't really witnessed how good LLMs can be for a subset of software engineering tasks. I find for mid-complexity code they are far, far better than I am at coding. Even though I often tweak the
550.
▲
by
zmmmmm
1y ago
It's why I use Aider, because it only operates on explicit files that you give it. Works great with OpenAI but if you are really worried, it interfaces perfectly with Ollama for local LLMs. A 12b model on my Mac does well enough for co
551.
▲
by
zmmmmm
1y ago
Seeing Meta sell millions of them?
552.
▲
by
zmmmmm
1y ago
I think the article subtley misinterprets the whole idea of it and the problem of it. It's fine to invent your own wheel. But that doesn't mean you should put it in production or feel any entitlement for anybody else to use it the
553.
▲
by
zmmmmm
1y ago
Oh man, squircles are back in fashion big time. They just waste so much space with their large border radiuses, it frustrates me.
554.
▲
by
zmmmmm
1y ago
That's really good to know as I have both a Quest 3 and a Quest Pro myself. What I'd say is that Quest 3 is about the minimum I can work with. I did work with Quest Pro for a while and it was just usable but only as an exception.
555.
▲
by
zmmmmm
1y ago
It would be nice if that was true but in my experience the vast majority got promoted for their success one level down and just found themselves lumped with those responsibilities.
556.
▲
by
zmmmmm
1y ago
I hope soon. I hate managing people. If a bot can do all the HR busywork that would be fantastic. Just now I'm recirculating for the 2nd time a quote for a laptop that admin sent back to me because it had a tiny detail wrong, and then
557.
▲
by
zmmmmm
1y ago
that's really good to know When you look at reviews they all exclaim how clear and crisp it is, but they are virtually never looking at text, which is approximately the only thing I want to look at.
558.
▲
by
zmmmmm
1y ago
Can you elaborate what the main issue is for you?
559.
▲
by
zmmmmm
1y ago
I have been eyeing off this setup for ages but I'm stopped by the fact that the glasses only project 1080p displays. They seem to have been stuck at that resolution for years and I'm not sure if it's a technical limitation or
560.
▲
by
zmmmmm
1y ago
Consider that in the mid 90's when it arose, most organisations were writing business code in C++ and paying license fees for things as simple as string handling libraries. There was almost no good solution to cross platform network co
561.
▲
by
zmmmmm
1y ago
The worst thing about EntityBeans is they were so bad they made Hibernate look good, which led people to think it was good. After 10 years of hammering against ORM complexity I finally switched to using thin database wrapper layers and ha
562.
▲
by
zmmmmm
1y ago
Amazing! Curious, what happens to performance? I assume you still pay the same performance price for longer context, even if you can now fit it in memory.
563.
▲
by
zmmmmm
1y ago
I don't agree directly with this, but a variant of it does bother me: will the auto-regressive nature of AI eventually limit the novelty of the ideas humanity can come up with? So many breakthroughs come from people who work either i
564.
▲
by
zmmmmm
1y ago
I think a better reason autism is a bad example is that part of its definition is that it is a consequence of fundamental brain structure and development (differentiating it from other psychological disorders which are acquired and more mal
565.
▲
by
zmmmmm
1y ago
My guess is that initially it will get eaten away by parties that currently trade in US dollar for convenience shifting to trading directly in their native currencies. It's a pure cost / risk equation for them that they have to ma
566.
▲
by
zmmmmm
1y ago
Which "smart" person advised him to calculate tariffs based on trade imbalance for each country, including uninhabited islands? It's an utterly stupid idea that makes no sense at all.
567.
▲
by
zmmmmm
1y ago
This seems superficial and doesn't really get to the heart of the question. To me it's not so much about bits and parameters but a more interesting fundamental question of whether pure language itself is enough to encompass and en
568.
▲
by
zmmmmm
1y ago
Sounds like there are a range of reasons, but the bigger picture explanation is : Google no longer cares about incentivizing apps to be on the store. The mobile OS wars are over: every company and dev that wants to do anything is locked int
569.
▲
by
zmmmmm
2y ago
first party cookies yes But third party cookies are a lot more insidious, because they get sent without any visibility to the user and have generally peripheral relevance to the application they are using. It's like if you go to the su
570.
▲
by
zmmmmm
2y ago
> It makes the system more reliable It might in theory but I suspect it's going to make things very very unreliable for quite a while before it (hopefully) gets better. I think probably already a double digit fraction of our infrast
More ›