RataNova

What stood out to me wasn't just the farming process (which is fascinating in itself), but how much of it sits at the intersection of tradition, family and constraints that don't quite make economic sense anymore

Nothing says "this is real engineering" like discovering that even the bathroom has a dedicated expert and a troubleshooting playbook

How much of aerospace design used to treat the crew as an adapter bolted onto the machine

A good reminder that standardization matters

When the problem is well-defined, the backend systems are integrated, and the AI has actual authority to act, it can be dramatically better than traditional support queues

I think the post is implicitly assuming a much narrower scope than "full service advisor" but it doesn't state that clearly

This is cool, but I wonder how it holds up outside the "happy path"

I think a lot of people are starting to feel this

Feels like the real missing piece is user control: let people opt into updates, choose what gets enabled, and turn off data collection

It's not just a "rich city" effect. That's kind of the key point in the whole EV debate... once it's mainstream and infrastructure is there, it stops being a luxury signal and just becomes… normal urban life, with some pretty noticeable side benefits

I don't think the author is saying "subscriptions are good", more like "if Honda isn't even building the capability, they're not even in the game"

Honestly this feels less like Honda "killing EVs" and more like Honda admitting it never really committed in the first place

I suspect a lot of people don't actually dislike software development, they dislike developing without anyone invested in their development

Learning the lower layer felt like earning access to the next level of reality. You had to understand the constraints to make anything happen at all. Now it increasingly feels like you can just describe the intent and skip straight to the outcome.

A lot of people aren't actually chasing leisure or money as much as they're chasing irreplaceability

In real life, every additional data point has a cost...

I don't think this is doctors being captured by the system so much as medicine being cautious about scaling interventions without strong outcome data

But medicine isn't quite the same feedback loop as downtime monitoringg

But saying MRIs "suck at cancer" feels off. They're actually first-line or gold standard for certain cancers

It seems like the key missing piece is long-term randomized data showing mortality benefit and cost-effectiveness in average-risk populations

AI doesn't fix design debt, it amplifies it

Retro emulators are a perfect "happy path" for vibe coding

Speedups without changes in expectations just reset the baseline, and then you're sprinting forever

Even with all the pain, getting real units into real people's homes is still kind of magical

In some sense it seems like hardware has two cliffs: the handmade one, and the factory one. You need to fall off both to really know what you’re doing

What China really has is not just cheap labor but dense, living industrial ecosystems

Feels like the real skill in hardware isn't just engineering, but learning how to be absurdly literal with reality

How many of the failures weren't engineering problems, but interface problems between humans: missing labels on drawings, ambiguous cable lengths, assumptions about powder coating thickness, etc

We seal things, dry them out, sterilize them… and life just adapts sideways

So routine cleaning only works for objects that can tolerate it, and a lot of cultural heritage simply can't