News for hackers. Stuff that matters.

abc42

Would be fun to have a so called job where you can spend your days doing something like that.

Getting a second off boot time is of course a pretty stupid endeavor but latency in general is a very worthwhile thing to optimize, even when the gains are in subseconds.

Re:Claude Ultra ($200): 20x -> 10x Pro usage every week on Saturday October 31, 2026 @06:15AM (#50314621)
It looks weird, but the 20x before this change referred to the 5-hour limit, not the weekly limit. They're removing the 5-hour limit altogether.

So the weekly limit is not changing, just the naming is less confusing now. And the title of this post is false and should be changed.

Re:Lula wins election after final votes tallied on Saturday October 31, 2026 @02:54AM (#50313831)
Lula's victory was a US psyops? That's some 4D chess.
Re:Lula wins election after final votes tallied on Monday October 26, 2026 @01:34PM (#50247671)
Oh, in this particular case they have no need, as both candidates are friendly to Russia.
Re:Lula wins election after final votes tallied on Monday October 26, 2026 @01:32PM (#50247621)
>I think this framing is pretty much disinformation itself.

Russian troll factories etc are a documented fact, not an opinion. It goes way beyond just "influencing elections".

Re:Lula wins election after final votes tallied on Monday October 26, 2026 @06:04AM (#50241463)
They are children compared to Putin in the art of geopolitical manipulation. If he didn't wield his skills so malignantly, I'd be impressed.
Re:Lula wins election after final votes tallied on Monday October 26, 2026 @06:04AM (#50241461)
>Why is it always Russia?

Why indeed.

Re:Lula wins election after final votes tallied on Monday October 26, 2026 @06:02AM (#50241449)
That's an interesting angle. Is that a fact, though? Far-right has usually been in bed with Putin.
Re:Lula wins election after final votes tallied on Monday October 26, 2026 @02:39AM (#50239931)
Interesting how pro-democracy wave after wave strike the shore as Russia's resources dwindle. Perhaps Russia is/was the only thing still keeping those monsters up?
Re:Struggling to Keep AI Out on Sunday October 25, 2026 @04:58PM (#50236003)
Why are we enduring this retrograde bullshit on a site like Hacker News? It was cute and perhaps even slightly poignant for a while, but it's starting to sound really really dumb.

Try https://lobste.rs perhaps. They love this kind of content.

Re:Digg on Sunday October 25, 2026 @10:31AM (#50231987)
To my personal shock, Digg is good now.

Highlights, imho:

1. aggregation of multiple sources into single posts

2. no comment sections

3. UI is pretty nice now

I wonder if it'll be ready before Trump is out.
I don't understand the question. Greed exists in every human system because it's a human quality, not a system quality. Real people are driven by many, sometimes conflicting things at the same time.

Non-greedy action is by definion an action that doesn't directly and materially benefit the one acting. That is universal across all political and I would argue all biological systems.

There are lots of drives besides greed.
Re:Roku developer repositories moved from GitHub to GitLab on Saturday October 24, 2026 @01:44PM (#50223459)
How about "if you're not the customer, you're the product".
Re:Roku developer repositories moved from GitHub to GitLab on Saturday October 24, 2026 @01:43PM (#50223447)
Why is this important news?
Re:Crystal 1.22.0 on Saturday October 24, 2026 @03:09AM (#50218932)
Cool. I'm about 3-4 months in, this is the current result: https://hanki-lang.org/docs/reference/ -- sorry about the LLM smell. I haven't paid much attention to making the documentation beautiful yet.

Seems like you're slightly more ambitious with the features (linear typing, refinement typing are way beyond what I dream about). I'm focusing more on a small practical application language with ruby-like aesthetics.

I wanted Hanki to be a general-purpose language that could be used with minimal compromises as a scripting language, a configuration language and a reasonably high-performing application / server-side language. I wanted a good REPL, arbitrary precision arithmetics by default, reasonable safety features (effects, strong typing, concurrency via message-passing actors, erlang-like supervising) and excellent readability for human reviewing.

I wanted to include features that force AI agents into a style, so the linter is included as part of the checker and every linter complaint is an error.

The implementation is currently 300K+ lines of Rust, about 60Kloc of Hanki.

I still have about a hundred beads tickets to go through before I'm happy enough about this that I dare to publish it. And seems like every solved task gets 2 follow-ups :)

Re:Say goodbye to manually typing replies on x on Friday October 23, 2026 @02:04AM (#50204998)
I support everything that makes X a more horrible place thus advancing its total collapse. So this is great.
Re:Getting ready to operate GNU Taler in Germany on Friday October 23, 2026 @02:02AM (#50204980)
I think the proper comparison point is not something like Monero, but multinational credit card companies. Taler is obviously much more private than those.
Re:Crystal 1.22.0 on Thursday October 22, 2026 @01:27PM (#50198427)
Crazy how easy it is these days, isn't it? The last 20% that matters is gonna be a bit harder though, I assume. Plus with everybody being able to do it means that making yours be interesting is going to be extra difficult.
Re:Getting ready to operate GNU Taler in Germany on Thursday October 22, 2026 @01:24PM (#50198369)
I think Taler is designed to be privacy-preserving while being able to comply with national laws. So not the currency you're looking for if you wanna buy something illegal.
Re:Crystal 1.22.0 on Thursday October 22, 2026 @11:02AM (#50196384)
They have an interactive mode right now (crystal i), but it's not very comprehensive. You cannot use certain libraries with it etc.

Shouldn't be terribly hard to make it good though. Some time ago, I added a JIT REPL in a few afternoons with Claude, ending up with roughly the same performance as their compiler, and with it working reliably with all the libraries I tested. But since they don't want LLM contributions in Crystal, I just left it as an interesting experiment and started working on my own language instead.

Re:There isn't an easy exploit for this PHP vulnerability on Thursday October 22, 2026 @10:58AM (#50196331)
>PHP is no more vulnerable than those languages.

In theory, perhaps not. In practice, much more vulnerable.

Re:PHP 8.2.35 on Thursday October 22, 2026 @02:32AM (#50191929)
Sure, but this particular vulnerability could not have happened in a language like Rust.
Re:There isn't an easy exploit for this PHP vulnerability on Thursday October 22, 2026 @02:29AM (#50191909)
That's a false dichotomy. PHP is mainly used in creating web pages, and a scripting language is not a requirement there.

If you compare it with production languages, PHP looks pretty weak.

Re:Bitbucket Is Down! on Wednesday October 21, 2026 @01:51PM (#50184927)
Atlassian stuff is Java, I believe.
Re:PHP 8.2.35 on Wednesday October 21, 2026 @01:50PM (#50184903)
That's how 90+% of script kiddy -type hackings happen, isn't it. That or publically available databases without a password.

AI is gonna make our stuff pretty robust, eventually. The transition phase might hurt a bit.

Re:DeepSeek: DeepSeek v4.1 Flash 1017 on Saturday October 17, 2026 @04:24PM (#50134786)
>And China is not really supporting Russia

Yes, yes they are.

https://en.wikipedia.org/wiki/China_and_the_Russo-Ukrainian_...

Re:DeepSeek: DeepSeek v4.1 Flash 1017 on Saturday October 17, 2026 @01:16PM (#50132978)
I support people making choices with full knowledge of what they are doing and what the consequences are, as much as it is possible. What I oppose is making choices without such knowledge.

Older ยป