News for hackers. Stuff that matters.

boltzmann64

Re:I Don't Understand Quantum Physics (2018) on Saturday October 24, 2026 @07:43AM (#50220348)
Saying I don't understand Quantum Mechanics is like saying, "I don't understand how attention/transformers explain the human brain."

It doesn't. It's model. All models are wrong (including quantum mechanics and attention) but some are useful. You understand the domain of the model, the problems it can solve, the math behind the models and use the model to make predictions. If you are competent at linear algebra and functional analysis, you can do quantum mechanics.

If you make the error of believing that quantum mechanics is ontological god's honest truth of nature, you will end up with blunders like using it to explain consciousness like Prof Penrose. His math was impeccable, but he was making a category error.

Re:The Future of CSS on Wednesday October 21, 2026 @08:20AM (#50179678)
Try tabularray.
Re:JPEG XL vs. AVIF for web use-cases on Saturday October 17, 2026 @11:45AM (#50132092)
so you fell for jxl propaganda. everybody who work with images for a living, knows that jxl has always been shit compared to avif.
Re:Show HN: My Terminal Themes Web App on Thursday October 15, 2026 @09:16AM (#50103224)
I used to hate font ligatures until I discovered Commit Mono (my all time favorite monospace font). They use the ligatures only to adjust spacing and make it look better on a monospace grid. No stupid <=, -> ligatures or anything else.
Can you please tell me what settings get auto re-enabled? I use Telegram as my primary messenger app. I just want to make a more informed decision if I should switch to Signal or something.
Re:Decisions API is in public beta on Wednesday October 07, 2026 @01:49AM (#49988784)
have you tried Moondream?

https://docs.moondream.ai/

Re:Release of Polars 2.0 on Tuesday October 06, 2026 @12:03PM (#49980483)
Learn SQL and interface with Duck. You will be 100x faster than Pandas/Polaris duo at fraction of memory. Also SQL is supported literally everywhere with a much more capable than Pandas API. Duck outputs to a Pandas Dataframe, but just treat that like a dictionary. Do all your processing, filtering and aggregation in Duck.

Also, try fx.wtf as a replacement for jq. it comes with a in-built tui viewer that supports vi-keybindings. Ecmascript is built into fx.wtf so you can query the JSON with JS notation (where JSON was born). You can use any JS functions including map/reduce/filter or perform any kind of transformation instead of learning jq dsl that you will forget tomorrow.

Re:AMD's random number generator can't generate a 0? on Tuesday September 22, 2026 @08:22AM (#49800072)
i remember some linux kernel dev got ousted by the community because he/she wanted to not implement a backdoor that would compromise the results of /dev/urandom.
this is is not new. we have had factory model (forking the initial window) in terminal emulator since 1990s.
Re:My Mathematical Regression on Tuesday June 23, 2026 @03:59AM (#48641780)
or just rotate it by 45 degrees and pretend it's a pascal's triangle.
Re:Splash Is a Colour Format on Monday June 08, 2026 @05:00AM (#48442897)
then, why does 222 look blue/green and not dark gray?
Re:Stop Using Conventional Commits on Friday June 05, 2026 @11:06PM (#48421003)
Wait. i am already struggling with 50 columns for the title, now i have add this `<type>[scope]:` bloat to it? How are these people doing both "Conventional Commits" and 50-72 at the same time?
Re:Why Janet? (2023) on Tuesday June 02, 2026 @09:44AM (#48370188)
if those are the reason why you love janet, then you will love tcl because you will be able to do all the same things without drowning in parenthesis and weird syntax.
I am really curious: what is your definition of a power user.
Re:Dithering with CSS on Friday May 08, 2026 @07:56AM (#48061833)
please add atleast one example to demonstrate how the output will look like.
how is this a dark pattern? say that you bought a subscription of new york times for a year and put it on monthly installment. can you bail out of paying the credit card company after six months? only here, both the product and credit card company is adobe. in fact adobe is being generous by letting you "cough up" only a half the fee where your credit card company wouldn't. at this rate all credit companies are "dark pattern peddlers" according to you? adobe is pretty clear on the name of the subscription, "yearly subsciption, payed monthly" in big letters at the top of their pricing page.
Re:Just Enough Chimera Linux on Tuesday April 14, 2026 @08:22AM (#47764681)
if you want a stable chimera linux as a daily driver, go to voidlinux.org. chimera linux started as a void linux fork until it became it's own thing. they share the same dna. cbuild started as a xbps-src fork.
Re:Undo in Vi and Its Successors on Tuesday February 17, 2026 @09:21AM (#47047761)
If you are going to need this, then you should use git instead of relying on the editor's undo tree.
Re:NPMX – a fast, modern browser for the NPM registry on Saturday February 14, 2026 @12:51AM (#47011966)
I hate the trend of moving away from being able to search packages from the command line to searching inside a bloated web browser. It had a happened with PyPI. And now npm. Please stop this madness.
Re:AI can 10x developers in creating tech debt on Saturday January 24, 2026 @01:03AM (#46741455)
Non-native speaker here. Is the phrasing of the blog title awkward or am I the only one? Seems like they are using "10x" as a verb and my brain kept parsing "10x" as a adjective to developer, reading "10x developer" which is a already established industry lingo.
This is the standard playbook. And the gov just pulled a switcheroo:

Policy that is hard to pass: SIM binding for all messenger apps and automatic log out every 6 hours for desktop apps.

Even more egregious policy: Pre-install spyware that cannot be disabled.

Withdraw the egregious policy on outrage, and people think they have won the battle.

Re:The State of GPL Propagation to AI Models on Saturday November 29, 2025 @09:08AM (#46087636)
Google's engineers when they were copying Java API for Davlik (and later ART), they had access to and consulted Java source code. The infamous Oracle v. Google judgement siding Google set precedent at the highest level, SCOTUS that looking at the code is not an issue.

So, it doesn't matter if a AI can or cannot do clean room implementation. Unless it is a patent or trade secret violation, cleam room implementation doesn't matter.

Re:Drop into REPL when your Python script crashes on Wednesday October 22, 2025 @11:12AM (#45670379)
Thank you for the usercustomize.py tip.
Re:Ripgrep 15.0.0 on Saturday October 18, 2025 @11:07AM (#45627972)
Has it caught up to ugrep in terms of backward compatibility and speed yet?
Re:Why Today's Python Developers Are Embracing Type Hints on Sunday September 28, 2025 @06:30PM (#45408679)
I am sorry, but whats wrong with doing something like, `print(type(var)); exit()` and just running it once instead of digging through 5-6 stack frames?
Re:Jujutsu for Everyone on Sunday August 31, 2025 @03:04PM (#45086030)
I am think you are misunderstanding the entire premise of git. There is no "force push to trunk breaks every other developers' working copy." There is no central repo/trunk where all the commits are pushed. You are probably thinking of svn.

In git, your repo is the canonical repo and that is where you work. You work on a new feature and when you are ready, you "git format-patch" and "git send-email" to the community via the mailing list or other developers. A discussion may happen and people may or may not decide to apply the patch to their own repositories, with "git am." This doesn't break every other developers' working copy because they decide how to apply the patches they got in their email. No central repo, no trunk, guaranteed by the d in git dvcs.

Older »