Praying that KD and COTY rise, validating my purchase of both at their lows.
Liquidating other bits so I can be flexible when my gold miner…AEM…announces Thursday pm
Datacenters in space are just such a hilariously bad idea from nearly every angle that it amazes me that anyone with even a basic understanding of IT or even logistics can't smell the BS. This should be driving investors away, not bringing them in.
First, the vacuum of space is one of the best forms of thermal insulation in existence. You can basically only shed heat by emitting IR radiation or shedding mass (i.e. heat syncs) and data centers are famously known for generating a lot of heat. Second, you are exposed to billions of cosmic rays and other charges particles as well as micro meteors and other hazards that will play havoc by flipping bits in memory or physically destroying your infrastructure. Third, you would need a perpetual manned presence to monitor and maintain the datacenter. Fourth, all communication to and from will be subject to severe, varying latency issues dependent on current distance from the orbiting data center. Fifth, it's a huge legal loophole due to the inherent neutrality of space. Who's data laws hold sway? And who will be able to go to space and confiscate the infrastructure if it is subpoenaed?
And all of that doesn't even mention the cost and difficulty of sending any mass (let alone delicate server infrastructure) into orbit and the threat of kepler syndrome from adding more junk to space.
Just bits a pieces. I was about 40% cash. Down to about 30% now. Buying some things I wanted but at a better price. Still kinda seeing where stuff goes.
This sums up exactly why data centers in space is stupid. Musk also yapped how he will go to Mars and other scientists just said how stupid that is . Even Neil deGrasse said that if we had the technology to do something on Mars we would have technology to do much better on earth. I am trusting astrophysicist over Musk any day
(So you would have heard the obvious news about SpaceX and X. Not convinced by the proposition really.
Okay, let's break this down because the idea of putting a datacenter into orbit sounds amazing until you actually look at how space works.
First, everyone pictures space as this freezing cold void, perfect for cooling, right? It's actually the opposite. Space is a thermodynamic prison. There's no air, so you can't just blow fans over hot components. All that insane heat from millions of processors has exactly one way out: it has to slowly radiate away as infrared light. To do that on a data-center scale, you'd need to build these gargantuan, delicate radiator panels. We're talking about a structure needing square kilometers of surface area. Like FFS imagine trying to deploy and protect a radiator the size of a small city. One analysis suggested a 5,000-megawatt facility would need about 16 square kilometers of combined solar and radiator area. For scale, that's hundreds of times bigger than the International Space Station's arrays.
And that brings us to the second nightmare: space itself is trying to kill your computers. It's flooded with cosmic radiation and solar particles that constantly barrage electronics, flipping bits from 1 to 0 and corrupting data silently.
* To fight it, you'd need either massively heavy shielding (which rockets hate) or
* you'd have to use specialized, slower, and way more expensive "rad-hardened" chips.
So you're either paying a fortune to launch a lead-lined server farm or you're not even getting top-tier computing power up there.
Then there's the orbital junkyard problem. Low Earth Orbit is already cluttered with debris - old satellite parts, flecks of paint - all zipping around at about 15,000 miles per hour. Your sprawling, kilometer-wide radiator complex would be sitting in a cosmic shooting gallery. A collision with a piece of debris the size of a marble would be catastrophic, potentially creating a cloud of fragments that could take out the whole structure.
But the real dream-killer is the sheer, absurd economics of it all. Let's talk launch costs. Even with reusable rockets, it's brutally expensive. At a rate of roughly $1,500 per kilogram, just launching a single, standard server rack (easily 1,000 kg or more) could cost $1.5 million... and that's before you pay for the actual servers, the solar panels, or the giant radiators.
The scale is mind-boggling. One estimate suggested that to replicate just 1% of Earth's total computing capacity in orbit, you'd need to launch over twice the total mass humanity has ever sent to space in history. The numbers just don't close. The capital required would be in the trillions, all to (maybe) save on electricity bills decades from now.
Now, is anyone even trying? Sure, in a very small, experimental way. Companies like Sophia Space are working on neat integrated tiles, and whispers of projects like Google's Project Suncatcher aim to send a couple of test chips up by 2027. Or even Starcloud, backed by YC. I think an Indian start-up was also there, TakeMe2Space, IIRC. But I'm not convinced.
The smart money is on solving those problems where they exist: better nuclear reactors, advanced geothermal, and just building data centers in cooler places on Earth. The orbital data center is a fantastic backdrop for a sci-fi movie, but for the foreseeable future, that's exactly where it belongs.) stolen
only software sector needs to worry about that, which basically all medium / small cap software companies get reduced to being labor companies. they become the circlejerk passing around tiny bits of money but the large caps all get to reap the cost saving benefits. when you live in an age which everyone has cheap software printers, everything is about hardware, or how captive your userbase is. the latter is dependent on the quality / stewardship of the company (this cannot be regulatory captured as it'd mean we'll begin importing all of our apps like tiktok) where as the form is dependent on what you already have access to. (SPY 1200 12/15 - but only because inflation lol)