[lapinot]

Any particular reason why you couldn't adapt energy consumption to match disponibility? 99.999% availability is a nice perk to have, but you have to overprovision to get to it or otherwise have big costs and complicated problems. OTOH making factories more lowtech, simple/robust/small-scale surely they're gonna have lower peak efficiency, but they'll cost less and be start/stoppable far easier. Scaling has limits.

[nickpp]

Changing the structure of our distribution network and the pattern of our energy consumption will cost much, much more (in both time and money) than building additional production into our current system.

[lapinot]

A couple nitpicks before contradicting your claim:

Time and money are both cheap in the face of climate crisis ;)

> than building additional production into our current system

Not sure how this is can lower energy consumption and emissions.

You say "our" distribution network. I'm quite positive the level of electricity availability most people in the world have access to is not the kind of thing we're seeing here (france). Even in us/eu/"highly-developped" countries, there is lots of variance between city and countryside. So there's a big part of us not making them build things like us that make them depend on us. Which would be quite revolutionary on it's own.

---

This is not something i'm making up, some detailed projections of such low-tech scenari have been done (eg "the age of lowtech" by bihouix, part 2). Most infrastructures are readily overprovided (network infrastructures generally like roads, electricity distribution, telco) and could be left as-is for long time if we lower the requirements. Ponctual plants can be either selectively shut down and working intermitently (stuff where scale is really enabling like some chemistry plants, primary industry or semiconductors) or scaled down and relocated (everything household, everything clothing, everything construction, everything food).

I fully agree that there is a very big inertia, as in infrastructures aren't fluid. But that not a fatality. Globalisation has benefits for some stuff (again semiconductors, petro chemistry), but for most of the volume (consumption goods) there's no material benefit (i'm not talking financial things, i mean from an engineering perspective) to globalize food, textile, consumer good manufacturing. For these stuff there are marginal process efficiency benefits at the (externalized) cost of maintaining very highly available network infrastructure (and obv making parts of the world less gifted by history dependent on the other).

In particular for consumer goods, not only does it not make any sense outside of wealth-extraction to globalize, but we're building too much and too cheap. There's no economic incentive to do it, but we know since long how to build stuff that last. They cost more to produce (bulkier/more material, less (more crude) plastic and more wood/metal), are less efficient on a small scale (thus require shared ownership) and have several order of magnitude longer lifetimes, etc. But we could literally flick a switch, stop production, repurpose/scrape the newly available high-tech plants and no one will be any wiser. And not only plants: we'd free up maritime transport, manpower, etc. We'd pay with reduced capitalistic output and trend-based lifestyle, but materialistic well-being would actually be better: we could all get access to what we call "professional" gear.

edit: your claim also seem to get contradicted by proponents of the smart-{city,grid}, right? All there needs to be is some more coordination. Which would also be easier to do with less and bigger things which live longer.

edit2: let's dream for a moment and imagine what could happen if we just took control of some big industrial infra with capital, domain knowledge and all and gave it 5 years to come up with public and durable goods for concrete tasks, sharing all the details and teaming up with the f/loss engineering world.