Programming in the Apocalypse

Maybe it's time to start putting the "Cloud" in orbit.

Some people may have understood, or rather suspected. They weren't in the majority though, or lots of misguided geoengineering wouldn't have been done the way it was.

Clearing forests has been progress thing until relatively recently, draining swamps was a totally great thing until even more recently, straightening rivers into concrete beds has been considered progress up into my lifetime. All of these things have had lots of bad downstream consequences to the point that lots of places now spend huge sums to undo at least some of these developments. And come to think of it, wildfire management is another geoengineering effort that lots of very dry places totally screwed up (e.g. California, as discussed frequently on this very site) out of the very best intentions, but working with broken models.

Plenty of times species have been introduced overseas to control other species, and it's always been a disaster, to the great surprise of everyone involved, time and again. Dams for hydro power are still considered ultra low ecological footprint power generation by lots of people, even though they form barriers that can completely disrupt river ecosystems to the point of leaving desolate wastelands in the riverbed downstream, disrupt riverside ecosystems downstream that depend on regular floods, allow for excessive water extraction and so on. Hunting predators to extinction is still widely popular, even though ecosystems without predators can't and don't function. I've had conversations with local people who're hellbent on exterminating the few remaining local beavers because they damage trees; but beavers are a keystone species, tons of species depend on beaver-created clearings.

The list doesn't end: Dumping toxic waste into rivers has been considered harmless until toxins accumulated to levels extreme enough to severely hurt people, by which point some of the worst-polluted rivers had been pretty m much sterilized (e.g. the lower Rhine). It's hard do believe this today, but people did honestly think nature would take care of the gunk, filter it out or dilute it or whatever. We pumped lead into the environment by means of leaded gasoline, one of the craziest "accidental" geoengineering adventures to date, until whole forests started dying, and of course some people saw that one coming, but then some people saw the world end when the LHC went online, and good thing we didn't listen to those people. When I grew up, climate change was widely considered a crazy myth; some saw it coming early on, the majority had a good chuckle; yet that's the biggest geohacking fuck-up in all our history, and it took us that long to realize the fact that climate change is real.

Generally speaking, with experiments like this, the true consequences tend to not become visible until way down the line, at which point cleanup may be impossible (e.g. climate change, the current mass extinction) so we need to anticipate such things and get it right the first try. Yet we've historically both failed to build non-rubbish models and then to heed those few warnings we did get. Convenience and progress and growth seem to always trump the naysayers, and often that's just fine – the world didn't end when the LHC went online and we learned a lot about the fundamentals of physics. Good thing we didn't listen to them.

But all this history leaves me personally highly pessimistic when it comes to more planet-scale climate hacking, given that we don't even understand the downstream consequences of our past and current climate hacking and given that our track record of getting this sort of thing right on the first (or any) try is so grotesquely bad.

We're great at problem-solving short-term, everyday issues with near-immediate feedback loops, like by mass-producing crazy good tools; we're bad when things get big, abstract, long-term, with long-ish feedback cycles e.g. when building nuclear reactors that don't malfunction in major ways, because we start making bad compromises and take shortcuts even though we should and do know better, because we socially can't help doing this; and we're sad failures at anything extremely large-scale, extremely long-term, extremely long feedback cycle-ish, like climate change or – planet-scale climate engineering.

Since we are bound to get pretty desperate and since climate hacking does offer an enticingly quick way out, I'm confident we'll try it at some point. When we do, I very much hope I'll find my pessimism proven wrong.

Edit: fixed formatting

Have you done any research about the impacts of climate change? Complaining about the urgency of warnings only works if you're certain the urgency is unwarranted.

I don't think any explanation is "better" or "worse" than the great filter, as we only have our example. There's nothing suggesting that we are alone in the universe, but also nothing suggesting that we aren't alone. We can formulate whatever hypothesis we want but it won't change reality. Either there's something out there or there isn't. Lots of people thinking the great filter or Drake's equation is real won't affect what is or isn't out there. We just don't know, and it's mostly a waste of time to think about it.

One should first ask what is the actually useful task for computers. Right now it is often things like powering ad networks, tracking engagements, running tax code for millions, calculating sha 256 hashes. Would any of this be useful in apocalypse? If not what would be?