Hacker News: larschdk

New comment by larschdk in "GitHub Copilot is moving to usage-based billing"

larschdk — Mon, 27 Apr 2026 17:43:20 +0000

There will surely be corresponding, different AI credit costs for each model.

New comment by larschdk in "GitHub Copilot is moving to usage-based billing"

larschdk — Mon, 27 Apr 2026 17:42:02 +0000

On top of being worth less, the subscriber discounts are gone.

The old plans were $0.033/request for Pro, $0.026/request for Pro+ and $0.04/request for pay-as-you-go. That discount is now gone. They even still advertise "5x the number of requests" for Pro+ over Pro.

New comment by larschdk in "OpenCode – Open source AI coding agent"

larschdk — Sat, 21 Mar 2026 07:49:45 +0000

No surprise that a tool that can run shell scripts, open URLs, etc. is flagged down on Windows where AV try to detect such trojan methods.

New comment by larschdk in "Floppy disks turn out to be the greatest TV remote for kids"

larschdk — Tue, 13 Jan 2026 08:28:16 +0000

An 125 kHz RFID reader would be a way simpler and cheaper solution. Could still have a 3D-printed box/slot.

New comment by larschdk in "FFmpeg has issued a DMCA takedown on GitHub"

larschdk — Sat, 27 Dec 2025 00:40:47 +0000

Copyright law defines derivative work by substantial similarity and dependence, not by technical mechanisms like linking. Technical measures such as linking is not a copyright concept.

Dynamic linking is a condition for LGPL compliance, but it is not sufficient. Dynamic linking does not automatically prevent a combined work from being a derived work.

New comment by larschdk in "What's New with Firefox 142"

larschdk — Tue, 02 Sep 2025 14:38:01 +0000

Here's a feature request: Disallow any and all unsolicited attention-seeking pop-ups or notifications from the browser itself.

New comment by larschdk in "SQLite (with WAL) doesn't do `fsync` on each commit under default settings"

larschdk — Sun, 24 Aug 2025 16:50:15 +0000

Just checked debian/ubuntu/alpine/fedora/arch docker images. All are FULL by default.

New comment by larschdk in "Yet Another Zip Trick"

larschdk — Sun, 06 Jul 2025 08:10:32 +0000

The Office365 online and desktop implementations of zip could be different.

New comment by larschdk in "JSLinux"

larschdk — Tue, 15 Apr 2025 11:01:22 +0000

Rather one long function than does one thing well than multiple function that are strongly coupled and difficult to reason about. Programmers who apply dogmas can be harmful.

New comment by larschdk in "Nvidia, ASML Plunge as DeepSeek Triggers Tech Stock Selloff"

larschdk — Mon, 27 Jan 2025 11:51:37 +0000

I think it is more expectation about expectation. You buy/sell based on whether you expect other people to expect earn or lose. It is self-referential, hence irrational. If a new play enters and peoples expectations shift, that affects your expectation of value even though the companies involved are not immediately or directly affects.

New comment by larschdk in "Who Can Understand the Proof? A Window on Formalized Mathematics"

larschdk — Fri, 10 Jan 2025 13:10:54 +0000

The source uses ○, not •, for the NAND operation.

New comment by larschdk in "Review of Mullvad VPN"

larschdk — Thu, 12 Dec 2024 10:38:12 +0000

Banknotes have serial numbers. Don't think that they are impossible to track.

New comment by larschdk in "Making memcpy(NULL, NULL, 0) well-defined"

larschdk — Wed, 11 Dec 2024 13:19:41 +0000

When C was conceived, CPU architectures and platforms were more varied than what we see today. In order to remain portable and yet performant, some details were left as either implementation defined, or completely undefined (i.e. the responsibility of the programmer). Seems archaic today, but it was necessary when C compilers had to be two-pass and run in mere kilobytes of RAM. Even warnings for risky and undefined behavior is a relatively modern concept (last 10-20 years) compared to the age of C.

New comment by larschdk in "Firefox-Passwords-Decryptor: Extracts and decrypts passwords saved in Firefox"

larschdk — Tue, 22 Oct 2024 10:45:11 +0000

Does this prevent other software running on the same hardware from accessing the keychain?

E.g. on Windows, any program can access the entirety of the credential store for the current user.

New comment by larschdk in "Show HN: X11 tool to share a screen area in any video meeting"

larschdk — Tue, 15 Oct 2024 12:56:27 +0000

Sure, if that is the only program, but it is not. This kind of thinking drains batteries faster than necessary, drains the cache, and reduces CPU efficiency. sleep() is a wasteful system call, a kludge at best, and is never the correct solution to a synchronization problem.

New comment by larschdk in "BorgBackup 2.0 supports Rclone – over 70 cloud providers in addition to SSH"

larschdk — Tue, 01 Oct 2024 06:50:49 +0000

rclone and restic are not direct alternatives. They have a slight overlap, but are also different. Rclone is more versatile for moving/copying files. Restic has snapshotting, pruning, client side encryption, deduplication, and compression. Restic actually supports rclone as a backend.

New comment by larschdk in "No more blurry fonts in Linux (2023)"

larschdk — Wed, 25 Sep 2024 06:30:31 +0000

Both use subpixel rendering is that it is specific to the display where it is shown. You need the same subpixel ordering and you can't really scale it. Your preference may be correct on your display, while mine is correct here: I strongly prefer the second image.

The first image has less hinting, causing the top of the x-height to be rendered at a half pixel. This makes it seem less intense that the rest of the letters. The second image aligns way better and have a more consistent intensity and sharp edges, but gives an overall slightly more bold appearance, and also compromises a tiny bit on the actual shape of letters.

New comment by larschdk in "Comfy, the 2D rust game engine, is now archived"

larschdk — Mon, 09 Sep 2024 12:03:44 +0000

I still like to believe that the upfront cost will be offset by the lower long term maintenance cost for some types of software. Except, that that is not always the case for gamedev where shitty code can still make good and successful games. You may miss a deadline, but if you are not going to maintain the code base for 20 years, so why spend the effort on quality code?

New comment by larschdk in "Posting a 5x5 crossword every day"

larschdk — Wed, 04 Sep 2024 06:14:44 +0000

Keyboard navigation needs some improvement. It's not easy to navigate because you are blocked from correct guesses. Also, it wanted me to guess "SASHA", but the hint was for "Sacha" (Baron Cohen).

New comment by larschdk in "Is My Blue Your Blue?"

larschdk — Tue, 03 Sep 2024 09:19:53 +0000

I am unable to answer many of them. I see mostly turquoise, not blue or green.