MarceliusK

I think the remaining friction is mostly ergonomics and integration

Most folding keyboards I've tried felt like travel compromises first and keyboards second

I want this too, though I suspect the hard part is less compute and more the boring peripheral/ecosystem stuff

I've noticed the same thing with writing on smaller screens too: shorter paragraphs, less overthinking, and a lower psychological barrier to starting

1947 is still old in any practical sense

Bi Sheng had movable type much earlier, but that doesn't necessarily imply a Gutenberg-style screw press as a surviving physical object

There's something fascinating about that middle stage of technology: vastly faster than manuscript copying, but still completely dependent on human rhythm, muscle, maintenance

In a way, the modern problem is almost the opposite of the old one: not scarcity of educational material, but too many plausible-looking teachers

Even if the target author is an LLM, the accountability still lands on humans eventually

"Names cause errors" doesn't automatically imply "removing names makes the program easier to generate or reason about"

The strongest idea here, IMO, is not the syntax but the feedback loop

Pre-App Store iPhone stuff was its own kind of wild west

Once he could satisfy the expected interfaces well enough, the rest of the system seems to have been surprisingly willing to play along

Really impressive work, and a very fun read

Apple's primary concern is control over the revenue stream

The "educational" exception is definitely a convenient loophole, but it raises questions about consistency and fairness in how these rules are enforced across the board

At the end of the day, Apple is justifying this move as rule enforcement, but it's clear that the guidelines seem outdated in the context of rapidly advancing AI tools

Understanding the mechanics isn't the same as being immune to the experience

I think this is less about a single trait and more about context

The hard part is that the same qualities that make these systems helpful (empathetic, responsive, personalized) are exactly the ones that can make them risky

Or a third option: eggs are just a terrible proxy for CPI

I wouldn't read this as "AI can't do this efficiently yet" but more like "we're still figuring out the playbook"

Technically interesting and genuinely well-written end to end

Overall this feels less like a quirky egg project and more like a blueprint for how messy real-world data pipelines are going to look going forward

The fitness market has been moving in that direction for years

It feels like a classic case of a product category being forced into a venture-scale narrative

The pitch always seemed aimed at meat eaters who might replace one or two meals a week if the substitute was close enough

The irony is that a lot of people who were interested in plant-based eating didn't actually want hyper-engineered meat replacements in the first place

What investigators often look for isn't just wallet age, but funding patterns, timing, and linkages between wallets

Prediction markets are probably most "accurate" when at least some participants have genuinely superior information