Hacker News: doph

New comment by doph in "Where the goblins came from"

doph — Thu, 30 Apr 2026 04:00:42 +0000

is a kv cache not a kind of state? what does statefulness have to do with selfhood? how does a system prompt work at all if these things have no reference to themselves?

New comment by doph in "I went to America's worst national parks so you don't have to"

doph — Mon, 13 Apr 2026 15:19:02 +0000

What I love about it is that more than any other park I've visited, it defies description. I adore Yosemite and Joshua Tree, for example, but I believe they can be understood from photography. Grand Canyon feels like it can barely be understood in person, like just the idea of it is too big or complex to hold in your mind and that is thrilling for me.

New comment by doph in "EDuke32 – Duke Nukem 3D (Open-Source)"

doph — Sat, 21 Feb 2026 21:12:59 +0000

I was into playing and modding Doom back in the 90s and just a few months ago rediscovered the community - I am just blown away by the effort and creativity that is going into these source ports and the indie games people are building on top of them. That passion and spirit of sharing is peak Internet/open-source.

New comment by doph in "A school locked down after AI flagged a gun. It was a clarinet"

doph — Thu, 18 Dec 2025 16:48:27 +0000

I think what's fair is to expect police officers to have as their top priority the protection the citizens they are serving. Too often they seem to have as a top priority the protection of themselves, which means as a consequence that they'd rather risk the lives of others than their own. What good are they to us then?

New comment by doph in "McDonald's pulls AI Christmas ad after backlash"

doph — Wed, 10 Dec 2025 16:18:47 +0000

There's a reason nearly all of the old guard VFX studios were driven to bankruptcy over the last decade and it doesn't have anything to do with massive demand for talent.

New comment by doph in "The Death of Arduino?"

doph — Wed, 19 Nov 2025 21:16:46 +0000

ESP32 - quite a range of dev boards and places like Seeed and Adafruit have a nice selection of accessories. Adafruit develops CircuitPython which is IMO the lowest barrier to entry for programming MCUs. Adafruit even has CircuitPython sketches on their site for how to interface with the components they sell.

Rust on ESP32 is still a bit early - the HAL crate is still pretty unstable, but the toolchain is quite nice and I'm able to be productive enough that I never reach for C or C++.

New comment by doph in "6B Miles Driven"

doph — Sat, 15 Nov 2025 16:46:08 +0000

I share this opinion. I've used FSD quite a bit during the free trial periods and each time come away with the sense that it's like driving with a newly licensed teenager at the wheel. If I have to be as alert and ready to avoid an accident as when I'm in command of the car, then this offers no improvement to the experience, just an added layer of stress trying to anticipate the actions of yet another actor in the environment.

New comment by doph in "6B Miles Driven"

doph — Sat, 15 Nov 2025 16:37:39 +0000

I'm in Los Angeles, which can be a challenging place to drive. Each time they give me a free trial of FSD for a month, I enable it and test it with excitement and optimism. Each time I only use it for a day or two before it does something dangerous enough to scare me.

New comment by doph in "Tiny electric motor can produce more than 1,000 horsepower"

doph — Mon, 03 Nov 2025 15:21:55 +0000

In-wheel application is possible, but it's important to understand that the pancake shape is only a consequence of the axial flux design and Yasa doesn't make motors in other "formats". Yasa motors shaped like this have been used in several supercars and all of them have been in-board on the axles, not in-wheel.

New comment by doph in "Tiny electric motor can produce more than 1,000 horsepower"

doph — Mon, 03 Nov 2025 15:00:07 +0000

I don't think their motors are axial flux, they're just large and narrow to fit inside wheels. Or at least all the images on their website depict radial flux designs.

New comment by doph in "My Impressions of the MacBook Pro M4"

doph — Fri, 31 Oct 2025 19:39:46 +0000

lots of people can notice that. my last job involved meticulously timing our software's input-tp-display latency, testing viewers' responses to it, and fighting for each and every ms we should shave off of it.

New comment by doph in "Wasp Blower"

doph — Fri, 24 Oct 2025 15:21:41 +0000

Very much this. After multiple very painful stings, I have a zero tolerance policy for nests on the house, but I am very grateful when they show up in the garden. Wasps are more effective at controlling garden pests than any chemical means I've tried. Plus they seem to be the only pollinators of my passionfruit.

New comment by doph in "Laptops create systems. Phones feed algorithms. The asymmetry determines power"

doph — Sun, 05 Oct 2025 14:29:23 +0000

Mostly agree, however this kind of quirk could issue entirely from post-training, where the preferences/habits of a tiny number of people (relative to the main training corpus) can have outsize influence of the style of the model's output. See also the "delve" phenomenon.

New comment by doph in "SpikingBrain 7B – More efficient than classic LLMs"

doph — Sun, 14 Sep 2025 15:58:14 +0000

Did they ever address that? I have not been able to stop thinking about it, it was so bizarre.

New comment by doph in "Simple design changes can make bat boxes safer"

doph — Fri, 05 Sep 2025 04:33:32 +0000

It hasn't been effective for me. I've put up two bat boxes, and neither has had any tenants in the 8 months they've been out. There are plenty of bats around, maybe they've got better housing options. The ones I observe occasionally fly low to the ground and near the house (it's a thrill to get buzzed by one), but mostly they're further away and higher up and we still get plenty of bugs, including mosquitos in the house. Bats probably do help reduce the numbers, but all it takes is one mosquito to ruin your night.

New comment by doph in "Claude finds contradictions in my thinking"

doph — Tue, 29 Jul 2025 15:29:31 +0000

The article you link is a very specific type of failure that apparently did not happen in this instance, where Claude was able to access the author's writing. And the author apparently found the insights useful, though the lack of analysis from the author on that value makes this article basically meaningless for an outsider.

I am apparently a different type of person than the author because my obsidian vaults look nothing like theirs, but I can't imagine asking an LLM for a meta-analysis of my writing. The whole point of organizing it with Obsidian is that I do that analysis myself - it is part and parcel of the organization itself.

New comment by doph in "How to Firefox"

doph — Tue, 22 Jul 2025 15:28:12 +0000

I didn't believe it and after trying those samples, I still don't. All of them run flawlessly for me on FF 104.0.4 on an up-to-date Arch install on my laptop.

New comment by doph in "Microplastics shed by food packaging are contaminating our food, study finds"

doph — Tue, 24 Jun 2025 13:57:57 +0000

Microwaving causes individual particles to join into a delicious plastic-cheese emulsion, making them undetectable.

New comment by doph in "If the moon were only 1 pixel: A tediously accurate solar system model (2014)"

doph — Fri, 13 Jun 2025 16:22:17 +0000

Yes, and this provides a nice intuition about the relation of wavelength to energy. But x and γ wavelengths are several oom shorter than visible light, so you'd have to be traveling at very close to c to experience that amount of Doppler shift.

New comment by doph in "Ocean Tides and the Earth's Rotation (2001)"

doph — Fri, 18 Apr 2025 13:44:22 +0000

>it's been about 100 years so now each day is 2.3 milliseconds longer

>after 1000 days 1000 * 2.3 milliseconds = 2.3 seconds

I don't think the example helps at all to explain the concept, but I think the math is right