eleitl @ eleitl @lemmy.ml

Posts

1

Comments

272

Joined

5 yr. ago

You might find your responses here a bit biased.

My first one was a random floppy distribution off an ftp server. I then tried SuSE off a CD-ROM and then went with Red Hat for a while. Then Debian, now Ubuntu. Tried FreeBSD but didn't stick. Probably Debian again next. I always prefer Debian on the server though.

Quarter century for me.

My primary consideration is all the expensive storage filled up by vapid image macros. 80 GB goes a long way for just text.

Pictrs should have been an optional microservice by default. Commenting here to keep track of this thread since this is useful.

Had it too. Uninstalling and reinstalling fixed it for me.

Many people run rented servers with capacity to spare.

Sign me up.

We certainly need a minimal logging mode for instances, and simple ways to comply with GDPR deletion requests.

Expect this to come to the EU in a few years.

Silent SMS are working as designed. There is a reason they are called silent.

So don't run large instances. Either selfhost at home if you're on symmetric fiber or use cheap ARM cloud instances.

You can't keep on track what is long over. At most you can return to Moore scaling, for a short while, but this is not telling you how exactly.

Don't forget lubricants, polymers and chemical feedstock in general.

There is biochar which can be scaled by providing cheap pyrolysis retorts and training how to bioactivate it. Incentive is increased agriculture productivity long-term. Capture should be done directly from flue gas since not needing enrichment. There are also carbon-negative concretes which have good potential for capture.

Synfuels from 100% renewable are at least carbon neutral, which is as good as it gets.

Sorry, I haven't started memorizing Fediverse handles yet. Tim Morgan measures ECoE in percent, and real GDP is GDP minus debt, or x units of debt produce a fraction of that in GDP. Plus ECoE accounting. His model is proprietary, so nobody exactly knows how it's computed but he himself.

That IRENA model seems to link some large spreadsheets and notes on that page. No idea how complete that is.

As to investments necessary, some two decades ago I estimated we'd need some 3 TUSD/year, inflation-adjusted, for the next 40 years to transition. As a rough estimate that's as good as any other guess, not facturing in extraction of progressively depleting resources.

The issue is more that we're currently running out of extractable fossils (net energy peak for oil liquids is projected to peak as early as 2025 and the decline into nonextractability is rapid) so it's a question of having liquid fuels and synthetic stock, at all.

What kind of life style we can expect where hydrocarbons and energy in general is expensive is an interesting question. See e.g. https://surplusenergyeconomics.wordpress.com/ for analysis of that.

About one order of magnitude more expensive than the fossil kind.

I'm only missing a hole in the wall to do the same -- and I could be running a 2x 40G link!

But the switch idles at 80 W and a starting jet noise level, so the hole in the wall will wait.

Nice. Where is power port on a hooker? Eh, nevermind.

I use photovoltaics, but not insular blackstart-capable battery buffered so when the grid is down I also have no power. Next year, though. Seems it's 2000 EUR minimum.