Seems to me that putting your datacenter in orbit might just possibly be the worst place for it, but what do I know.
IIRC data centers generate a lot of heat and, non-intuitively, it is difficult to get rid of heat in space. This is why the Space Shuttle opened its cargo doors when it got into orbit. They acted as a large radiator. Without them the Space Shuttle would have overheated inside the cabin.
I would think that is a much bigger problem for a data center.
That’s exactly right.
Not only the heat, but the non-reparability.
My wife showed me that link, and I said “anything that can be done on the ground almost certainly should be done on the ground. Moving it to space will only make it more expensive.”
Data is better down where it’s wetter, take it from me.
And power. Power is a lot cheaper on the ground, too. Sure, sure, solar panels… The only reason why spacecraft use solar panels is because they have to, because they don’t have any alternative.
Is no one thinking about how this makes it so much easier for aliens to steal our data?
I mean, they don’t have to land, learn oddly-formal English, steal wide-breasted suits to disguise themselves as Average Earthlings, kidnap an IT professional and fake his retinal scan to break into his workplace, look for the secret server farm, find out their IT dweeb has the lowest-level clearance, build teleport towers spaced perfectly around the computer complex, try to siphon enough power from the nearby city to make huge purple arcs of Van De Graf lightning that crackle without worrying the townspeople… except stalwart motorcycle mechanic Jake and his crush Gwendolyn the hacker whiz, who try to convince the local sheriff that the strangers in zoot suits are acting suspiciously, but now the computer complex is starting to flicker in and out of this dimension, but wait, is that Jake’s book group on their Harleys creating a diversion while Gwendy uses code from the original BATTLEZONE arcade game to prop open a back door into the aliens’ teleport program?
At some point power costs more than launching the payload. If they estimate that to be true around 2030, then it makes sense to start the project now. These sound like they are intended to be disposable satellites.
I think the real goal is to have a project framework to develop several technologies that they think they need in their roadmap. That and some ‘arrogance’ that they have to do these moonshots.
Sounds nuts to me. According to the article from Google Research, optimistically by the mid-2030s such a space-based facility may not cost much more than a ground-based one. So what is the point of doing it?
I’ll have what he’s having.
I wonder what I’m having. I read “retinal scan” as “rectal scan”, and I’m like, “Dayum! People will put up with anything in the name of security!”
Ahem… Radioactive isotopes and Peltier devices would like a word with you, sir.
(To be fair, unless you are (1) far away from the sun and (2) don’t need much power, solar panels are almost always the better choice)
And if you want to have some nightmares tonight thinking about all the ways it could go wrong, keep in mind that we have put nuclear reactors in space before. IIRC, we only did it once, in the 1960s…
The Soviets may have done something similar, not sure.
Wasn’t the original Thermos just a container within a container with a vacuum gap between them? Space may be cold, but the vacuum of space makes a really good insulator.