Hacker News: taneq

New comment by taneq in "Report on an Unidentified Space Station – J.G. Ballard (1982)"

taneq — Fri, 12 Jun 2026 16:52:28 +0000

I was assuming it was actually all set inside IKEA. :P

New comment by taneq in "Web Browsers on Video Game Consoles"

taneq — Fri, 12 Jun 2026 05:00:25 +0000

No idea, this was in the early 90s so we're talking very low resolution pics. I imagine they'd been sneakernetted around school yards for months on floppies after someone found their parent's stash and managed to yoink a copy.

New comment by taneq in "If you are asking for human attention, demonstrate human effort"

taneq — Fri, 12 Jun 2026 04:33:04 +0000

It’s much more effort to verify that code is correct than it is to produce it. This is the case even for human-written code, and now that we face a torrent of ok-looking probably-usable AI generated code, the problem is compounded infinitely.

If someone’s using AI to generate a large quantity of actually-tested, actually-good code then that’s one thing. If they’re generating a fire hose of slop and demanding that others do the actual human time-consuming work of validating that code then that person is the problem. It’s hard to tell which is the case here.

New comment by taneq in "Pokémon Go Scans Trained the Navigation Tech for Military Drones"

taneq — Thu, 11 Jun 2026 12:10:56 +0000

I’m so torn between naked admiration for the sheer Machiavellian audacity of this play, and discomfort with how vulnerable everyone is to this kind of creative abuse.

New comment by taneq in "Web Browsers on Video Game Consoles"

taneq — Thu, 11 Jun 2026 12:04:22 +0000

> I did a bit of curious searching on the family PC, but one time I forgot to wipe the history, and the game was up.

Hahaha that takes me back to the time my friend got hold of a 3.5” floppy with some Playboy pics on it, and then called me in a panic because Windows had helpfully added them to the “recent files” list and he couldn’t figure out how to clear it.

New comment by taneq in "AI agent runs amok in Fedora and elsewhere"

taneq — Thu, 11 Jun 2026 06:51:18 +0000

People without guns kill a lot fewer people than people with guns. Claiming that acknowledging this fact means you’ve been “driven mad by propaganda” is dumb.

New comment by taneq in "Siri AI"

taneq — Wed, 10 Jun 2026 02:47:23 +0000

AGI will either remove that need, or build its own factory to supply better phones cheaper than we could.

New comment by taneq in "Ultrafast machine learning on FPGAs via Kolmogorov-Arnold Networks"

taneq — Wed, 10 Jun 2026 02:45:03 +0000

You might be conflating throughput with latency. 100k tok/s is very different to 1 tok/10us.

New comment by taneq in "Siri AI"

taneq — Tue, 09 Jun 2026 08:50:24 +0000

I’ve wondered for years if this is what’s happening. They saw the mad scramble and hype bubble when gpt3 was released, and decided to sit it out until LLM tech was a bit more mature and commoditised.

Huge call if so, given that missing the bus on AGI (if AGI happened) is a universal existential risk, but it turned out to be the right one.

New comment by taneq in "OpenCV 5 Is Here: The Biggest Leap in Years for Computer Vision"

taneq — Tue, 09 Jun 2026 08:28:59 +0000

They’re multimodal LLMs trained for image generation. Turns out that if you want to generate images you gotta know what things look like.

New comment by taneq in "I'm building a parallel internet, and it's called The Thinnernet"

taneq — Tue, 09 Jun 2026 05:48:03 +0000

I’d say Markdown meets every one of those criteria except “more features” (which is a feature IMO) and “easier to monetise” (which is another feature. :D )

New comment by taneq in "MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second"

taneq — Tue, 09 Jun 2026 02:19:34 +0000

Therein lies the rub, no? To accurately predict the next token produced by a process, it’s necessary to model that process. If the process is a human attempting to estimate the duration of a task, then in some sense the LLM is modeling the estimation process. We’re well past the point where it’s credible to claim that LLMs just regurgitate their training data.

New comment by taneq in "Ask HN: What are tools you have made for yourself since the advent of AI?"

taneq — Tue, 09 Jun 2026 01:53:51 +0000

You can do interesting things with microwave kilns these days, I wonder if they get hot enough for ceramics? They can melt copper, I believe, so they’d be in the ball park.

New comment by taneq in "Ask HN: What are tools you have made for yourself since the advent of AI?"

taneq — Tue, 09 Jun 2026 01:53:05 +0000

I’m fixing up an old Bridgeport-style 5hp mill and converting it to CNC. I got it working enough to make myself a fly cutter out of scrap steel. :)

New comment by taneq in "OpenAI Submits S-1 Draft to SEC"

taneq — Mon, 08 Jun 2026 22:58:41 +0000

Just depreciate their server farms less this year to reduce losses. ;)

New comment by taneq in "Why are cells small?"

taneq — Mon, 08 Jun 2026 22:08:00 +0000

Why are things the way they are? Because it works better. Simple, really. :D

New comment by taneq in "Apple reveals new AI architecture built around Google Gemini models"

taneq — Mon, 08 Jun 2026 21:47:56 +0000

> Apple doesn't think parties other than themselves should be trusted with those iOS permissions.

I think we have ample evidence that regardless of whether Apple in particular is to be trusted, tech companies by default are certainly not.

Opening up access to users’ private data requires not just any given app to be trustworthy, but all of them.

New comment by taneq in "The 29th International Obfuscated C Code Contest (IOCCC) 2025 Winners"

taneq — Mon, 08 Jun 2026 11:57:20 +0000

Considering that one historical entry (winner maybe?) consisted of a source file that simply contained the character ‘c’, plus a makefile, I’d say it’s in the ball park.

New comment by taneq in "A Matter Wi-Fi Light Bulb in Rust on the Raspberry Pi Pico 2 W"

taneq — Mon, 08 Jun 2026 05:20:46 +0000

Use a digital storage scope, many of them can directly decode protocols like SPI . Then a lot of trial and error to figure out when exactly the issue happens and what else is happening around that time.

New comment by taneq in "The back cover of C++: The Language raises questions not answered by front cover"

taneq — Sat, 06 Jun 2026 06:59:31 +0000

This post discusses the topic and makes several key observations.