Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
pornel
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
91.
▲
by
pornel
4mo ago
Even if this flops, it's still better to lose money on this than the Metaverse.
92.
▲
by
pornel
4mo ago
It's so great that the US is against AI regulation and gives corporations freedom to innovate /s
93.
▲
by
pornel
4mo ago
Has extra KV cache on SSD, and lots more options to tweak. There's experimental TurboQuant and multi token prediction support.
94.
▲
by
pornel
4mo ago
Because as with every AI, when it passed, the goalposts have been moved. Naysayers say it doesn't count, because winning once isn't enough, it needs to win 50%+ of the time, or winning against an uneducated human who can't as
95.
▲
by
pornel
4mo ago
git repos viewed through gitlab's slow Ruby monolith are still git repos, so it doesn't make any difference.
96.
▲
by
pornel
4mo ago
64GB VRAM is too little to run good coding models IMHO. May be useful if you need voice models or run some slightly-smarter-regex batch processing or RAG workflows. Perhaps you're supposed to buy 4 or 8 of these and split inference acr
97.
▲
by
pornel
4mo ago
[meta] I wonder why people have such wildly different bar for what is "good" agentic coding? In a way, it's absolutely amazing that we've went from "Playing 'Set a Timer' on Apple Music" intelligence
98.
▲
by
pornel
4mo ago
This doesn't change a thing, because Rust is an answer to Modern C++ still being inherently unsafe and not meeting the bar. The best that C++26 has to offer is below the basic guarantees that Rust shipped in 2015. This is the blind spo
99.
▲
by
pornel
4mo ago
Incremental/quantitative improvement isn't disproven by contradiction. Anecdote is not data. Look at a bigger dataset, e.g. https://github.com/rust-fuzz/trophy-case there's a ton of overflows that result
100.
▲
by
pornel
4mo ago
Additionally, integer overflow is less immediately dangerous in Rust, because buffer access is bounds-checked after the arithmetic. You can still get some logic bugs that eventually lead to vulnerabilities, but it's not an arbitrary
101.
▲
by
pornel
4mo ago
There are recycling and trade-in programs that could collect compatible phones and pass them on in bulk.
102.
▲
by
pornel
4mo ago
This will be very controversial in the USA due to "or gal" being gender-inclusive, splitting Republicans and Democrats over whether the Palantir-Raytheon fast-track bill should say PPOBG or PPOBGOG.
103.
▲
by
pornel
4mo ago
As someone from another continent, I giggle when I see doomsday-prepping anti-govmint fiercely-independent cowboys hating EVs and loving gas. Gas has a 6-month shelf-life, and is attached to a whole geopolitically volatile military-industri
104.
▲
by
pornel
4mo ago
This problem is solved with infrastructure and legislation. It's something that improves almost overnight - your situation can change from being stuck without a place to charge to having it completely solved. All it needs is chargers i
105.
▲
by
pornel
4mo ago
In the EU case, Apple weaponizes people's ignorance about regulation. Apple pretends that the features everyone else has been shipping left and right somehow need extra paperwork and special approvals, because (…checks notes…) pro-priv
106.
▲
by
pornel
4mo ago
I just get an endless loop of Google captchas. They seem to have misconfigured CORS that prevents the CAPTCHA from reporting success.
107.
▲
by
pornel
5mo ago
> the author never actually tries agentic coding in Rust vs Python > Can't say I've tried Rust ... Well, I've actually tried both. The result is similar to what happens when humans code: for small programs the simplicit
108.
▲
by
pornel
5mo ago
Although the post focuses on RGB, the same quantization issue exists for any type of signal being mapped between discrete and continuous representations. The issue isn't in having a representation for 0 photons, but about maximizing in
109.
▲
by
pornel
5mo ago
No, the author understands the problem way deeper than you do. You haven't grasped the fact that the choice isn't obvious, and has subtle trade-offs. If you don't believe the author, check the other posts he references.
110.
▲
by
pornel
5mo ago
Such dismissive quip is unkind. The job has changed from a craft to operating an unreliable machine. Instead of satisfaction of solving challenging problems with your own skill and creativity, you babysit a text extruder and slog through mi
111.
▲
by
pornel
5mo ago
I don't think there's a consensus about that, as demonstrated by divided opinions on EU DMA and Apple vs Epic. The anti-regulation arguments aren't framed as "market competition is bad", but rather "the market
112.
▲
by
pornel
5mo ago
Unlikely, since library interfaces need to use a trait to accept all of the open/closed inclusive/exclusive syntax variations. If the only accept one specific named range type, they're clunky already.
113.
▲
by
pornel
5mo ago
The commands they list are app management, not part of LLM context. It's a bugfix for a needlessly delayed UI, not a model capability.
114.
▲
by
pornel
5mo ago
Rust's sync functions can block and await async functions. Which is another problem with the article: it doesn't clearly define what counts as having the "color". The problematic dead-end situation exists in JS, but lang
115.
▲
by
pornel
5mo ago
There are so many it's a meme: https://ourincrediblejourney.tumblr.com
116.
▲
by
pornel
5mo ago
Two things: scale, and humanity. People can't memorize as much information, and can't manually reproduce the works as quickly. There's a natural limit to how much damage a person can do without help of machines. That's w
117.
▲
by
pornel
5mo ago
I'm not sure about this, because a smaller battery pack with fewer modules needs to work much harder. PHEV batteries are discharged at higher rate (C) while a BEV can spread the load across many more modules. EV range of PHEV is so sma
118.
▲
by
pornel
5mo ago
OTOH asm.js can be retired now thanks to being backwards compatible with plain JS. It allowed it to be an experiment that could have been quickly rolled out without a risk of forever lingering as a back-compat requirement for browsers.
119.
▲
by
pornel
5mo ago
You can't mix them. It won't pass validation, and will fail to be optimized as asm.js. In asm.js you have to treat JS functions and objects as special extern values, just like in WASM. asm.js - when validated and optimized - is cl
120.
▲
by
pornel
5mo ago
LLM generation is bottlenecked by RAM bandwidth and latency. You can get almost linear scaling by evaluating more prompts in parallel, because the GPU has nothing to for the relative eternity it takes to read all of the weights from DRAM fo
More ›