It's nice to see old laptops get a new lease on life this way, but it's hard to deny the compromises that come with such an approach. As nice as the old 12-inch Powerbooks/iBooks, T60s, etc were, they leave a lot to be desired by modern standards.
This brings me to wonder how difficult it would be to develop a notebook that is specifically designed to be a focused offline ultraportable, using the most power efficient and cost effective components available along with an extremely lightweight, nearly nonexistent OS (think classic Mac OS) with some modern affordances and high-end touches (USB-C charging, milled single piece enclosure, etc). The imagined result is a machine that costs less than a decent Chromebook while being more responsive, having dramatically better battery life, and being generally more pleasant to use.
How would such a computer meaningfully differ from either a Pixelbook Go or an iPad with a keyboard? They both have stupefying battery life and there's not many ways to extend it because at this point virtually all of the power goes into the screen.
One of the benefits touted by the linked article is the forced offline aspect. The other is the much more restricted scope of apps, which keeps the user focused.
These are technically achievable with an iPad or Pixelbook, but require effort and discipline on the part of the user and on some level directly contradict the design and intended use of both products.
I think there are still significant savings to be had in terms of power usage/battery life. Lower power hardware puts a hard cap on how much any one application can consume and incentivizes efficiency on the part of the developer. The device's screen panel can also be selected for efficiency over density, brightness, color gamut, etc since those aren't priorities.
Arguably, these goals in terms of battery life / efficiency are already heavily incentivized for mobile developers - and by extension the iPad despite having "high end" processors.
For mobile devices, generally, the strategy is hurry back to sleep, IE ramp up the CPU to do a job quickly then put the processor back in the lowest power state, and for many applications this works great.
For the remaining applications, say mobile games, we might be limited more by the users desires (the best graphics, the lowest latency, the best colors, the most immersive sound) rather than this sort of hardware not being available.
All this to say, I agree with you that such a device could still exist, I'm just saying we might be that far from it.
It depends on what kind of sacrifices your willing to make, but if you really focus on low power you can build a laptop with multi day battery life. The upside of this is you end up with fewer charge cycles which increases battery lifespan in year 3+.
I'm using a T60 somewhat regularly and you'd be surprised how useful it still is.
Of course, a shitty "modern" website will perform exactly as you'd expect, but I try to avoid those even on my workstation so no big loss there.
Aside from that, it's perfectly usable for things like writing code, reading/writing emails, chatting on IRC, browsing lightweight websites like HN, reading documentation... you know, 99% of the stuff I use a computer for.
I have more powerful laptops as well, but I often reach for the T60 anyway, because it makes little difference for these kinds of tasks, and I love the 1600x1200 screen.
This brings me to wonder how difficult it would be to develop a notebook that is specifically designed to be a focused offline ultraportable, using the most power efficient and cost effective components available along with an extremely lightweight, nearly nonexistent OS (think classic Mac OS) with some modern affordances and high-end touches (USB-C charging, milled single piece enclosure, etc). The imagined result is a machine that costs less than a decent Chromebook while being more responsive, having dramatically better battery life, and being generally more pleasant to use.