Hacker News: cma

New comment by cma in "I requested a copy of my data from McDonald’s loyalty program"

cma — Fri, 14 Aug 2026 02:57:10 +0000

Still doesn't look like it is hurting them, but to put it in perspective that's a 4% increase in profit, while 2025 inflation was around 2.5%. Closer to a 1.5% gain in real terms.

New comment by cma in "Thanks to social media, canned sardines are a scarcity on the supermarket shelf"

cma — Thu, 13 Aug 2026 05:43:34 +0000

>We have apparently reached the stage

There used to be only 3 TV channels or so. You had things like Johnny Carson jokes causing a toilet paper shortage.

In Britain, Delia Smith's How to Cook caused 10% swings in egg sales, far more abundant and harder to affect than something like inventories of sardines. https://www.bbcamerica.com/blogs/british-icon-of-the-week-de...

New comment by cma in "YC startups are abandoning .com"

cma — Wed, 12 Aug 2026 15:16:36 +0000

"A recent YC startup spent 250k on a domain. I was slightly shocked. But when I asked how much they raised after YC, the answer was 6.5m. So 1/26 of their round. This is the time we live in; one shocking number counterbalances the other."

https://x.com/paulg/status/2083628660708561075

Was that a .xyz?

New comment by cma in "How Claude marks AI-generated content"

cma — Tue, 11 Aug 2026 23:42:16 +0000

When you tell the AI: copy this function to here, a small window rewriter would mean it just corrupts and changes it instead of moving it. And even it's own tool use would have some small window dumb model changing the tool calls based on what it thinks are synonyms? The Aaronson approach is much better than this, it's essentially like changing out the random seed for the sampling parts that were already random. For an operation like recall of previous text, the tight logits that result still keep it doing that close to deterministically. For something it creates itself, with more spread out probability mass, it gets watermarked.

New comment by cma in "Exploring Claude/GPT Knowledge Cutoffs and Pre-Training Timelines"

cma — Tue, 11 Aug 2026 05:31:08 +0000

Distilling is only prohibited from chats you prompt. But there are lots in the open like the LMSYS Chatbot Arena data. It's a essentially one of the largest public sources of preference data, maybe still useful for boostraping RLHF and successor techniques (possibly a more highly educated contributor base than the low paid contractors), and was full of GPT-4-era inputs.

New comment by cma in "Mark Zuckerberg attacks 'closed' AI rivals as Meta returns to open models"

cma — Mon, 10 Aug 2026 20:29:55 +0000

> The reason why llama is open source is simply because it got leaked

It arguably didn't really get leaked, and they had the .edu req mainly for fair use education exemption when legality of models was much more uncertain.

New comment by cma in "Taxi drivers rarely die of Alzheimer's"

cma — Mon, 10 Aug 2026 12:41:52 +0000

*meant to say conscientiousness (keeping brakes, tires maintained), not consciousness

New comment by cma in "Taxi drivers rarely die of Alzheimer's"

cma — Mon, 10 Aug 2026 04:27:02 +0000

Controlling for age in a profession with a high fatality rate like this seems like controlling for high alertness, consciousness, quick reaction times, and well restedness/low stress, low recreational drug and alcohol use.

Of those, sleep issues are highly correlated with Alzheimer's and also with auto accidents. Controlling for age alone here would select for low incidence of sleep issues.

New comment by cma in "Jeff Dean leaving Alphabet"

cma — Wed, 05 Aug 2026 16:46:59 +0000

Alex Turner discusses a bit here at this timestamp:

https://www.youtube.com/watch?v=pGlJQHVeKIA&t=14m27s

New comment by cma in "The applause that disrupted nothing – yet still led to an arrest"

cma — Wed, 05 Aug 2026 11:42:13 +0000

They get a good bit more disproportionate per capita representation in the senate than in the house.

New comment by cma in "MiniMax H3 Day-0 Support in ComfyUI: Open Weights, Native Audio, and 2K Video"

cma — Tue, 04 Aug 2026 10:01:57 +0000

This isn't the same with lookup tables, but in Explaining Attention with Program Synthesis they were able to replace some attention heads with Python programs:

https://arxiv.org/abs/2606.19317

New comment by cma in "Bitcoin cold-wallet attack spreads to 4,500 addresses as losses near $89M"

cma — Mon, 03 Aug 2026 04:11:05 +0000

Claude has found some weaknesses in other schemes, but not fully breaking anything yet:

https://www.anthropic.com/research/discovering-cryptographic...

New comment by cma in "Dubious research tied to Red Bull has shaped energy drink policy"

cma — Sat, 01 Aug 2026 04:43:23 +0000

That's if they are in a cave experiment: without that the sunlight cycle regulates it back to 24hr. But not for everyone, and that's N24/non-24, not DPSD, delayed-phase sleep disorder.

New comment by cma in "Kimi K3 Architecture Overview and Notes"

cma — Wed, 29 Jul 2026 10:27:01 +0000

The paper says:

"Kimi K3 uses no explicit positional embedding (NoPE), and instead encodes positional information implicitly through the recurrent gating and decay mechanism of KDA."

New comment by cma in "Substack writers, you need a website"

cma — Wed, 29 Jul 2026 10:14:35 +0000

Great way to not get subscribers to your substack etc.

New comment by cma in "Substack writers, you need a website"

cma — Wed, 29 Jul 2026 05:17:09 +0000

x.com heavily penalizes off-site links.

New comment by cma in "Kimi Linear: An Expressive, Efficient Attention Architecture (2025)"

cma — Tue, 28 Jul 2026 15:55:04 +0000

If they can distill fable into a full model post training run in ~15 days without the real thinking traces, yet we know Claude chats degraded with the thinking traces removed (chat resume bug from earlier in the year they reported stripping thinking to shed load as being the cause of degradation), how big can this degree be?

New comment by cma in "Harddrive Is Probably Full"

cma — Sun, 26 Jul 2026 00:26:10 +0000

Won't modern wear leveling migrate rarely written data to spread new writes out to its blocks?

New comment by cma in "Fil-C: Garbage In, Memory Safety Out [video]"

cma — Sat, 25 Jul 2026 08:04:05 +0000

You could use raw C in environments with effective sandboxes, Fil-C in environments without a sandbox.

Or you could write a library that works for Fil-C and C without writing the library twice.

New comment by cma in "Writing by hand is good for your brain"

cma — Thu, 23 Jul 2026 22:59:45 +0000

Where does this explanation fall apart? There must be a cutoff point where more difficulty of getting the info down is worse than more ease. Otherwise, wouldn't a better recommendation be to do all your notes more slowly tracing with your big toe in the sand and then taking pictures? Do touch typists get less out of digital note taking than hunt and peck typists?