A combination of solar furnaces (mirrors and a steering mechanism to track the Sun), and thermal depolymerization ( https://en.wikipedia.org/wiki/... ) can break down most anything organic into crude oil type feedstocks. That includes items like paper and animal byproducts. There is lots and lots of feedstock buried in landfills. So to reboot that part of civilization, you can use those ingredients.

In the aerospace industry, we had metadata around the actual design documents, and a process for incorporating changes. Some examples are:

* A drawing tree. A complete airplane or other complicated product had a top level drawing, that called out major assemblies (wings, landing gear, engine installation, etc). The major assembly drawings then called out sub-assemblies, in a tree structure, until you get to the parts level. Documents tied to a particular drawing (like engine installation procedure) got the same number as the drawing with a -002, -003, etc added, so you could track what they go with.

* Interface drawings and documents. Between assemblies you defined the interfaces between them - mechanical, dimensional, electrical, etc. You can't change your side of the interface before first consulting the people on the other side, and updating the interface data. That's how you ensure the pieces go together later.

* Requirements tracking. For example, the 747 landing gear has to support a takeoff weight of 880,000 pounds. Therefore there has to be a weights tracking process that assigns weight budgets to the various parts, and reports status back up the tree. Otherwise you can end up with a plane that's too heavy for the landing gear. Anywhere else there is a critical design value with contributions from various parts, you use this method.

All this metadata has to be passed around along with the actual parts drawings and software code. If you don't, then anything too complicated for one person to design is likely to need rework when the pieces of the design are merged.

Most makerspaces are hobbyist-level workshops. They don't usually have industrial grade software or fabrication machines available, because those are expensive. I'm working on the idea of a "MakerNet", where instead of a converted warehouse space with hobbyist tools and home-made workbenches, you have more commercial-grade machines spread around, either run as small businesses, or owned by groups of more serious hobbyists. For example, a $6,000 lathe might be split among half a dozen people. When you have a more serious project to do, you send the files for the various pieces to the respective machines that can make them. You also send payment, or deduct from a network account, to pay for the raw materials and other items you use up.

So higher quality machines, and people who regularly use them, therefore better output. But networked and distributed cost, so it is affordable on a hobbyist budget, and you have access to machines you can't afford on your own. Makerspaces can certainly be part of such a network. They would just need to have some machines and people that are able to do the better quality work.

Apparently moisture farmers and 'vaporators can be a real thing on Mars.

Perhaps the water is being insulated by the eddy of plastic trash in the central Pacific?

That's what bank safety deposit boxes are for. Offsite, hard to break into, more or less fireproof through sheer mass, even if the building around it burns. Ask the bank about how thick the walls are, though. Class 3 is recommended (12 inches thick concrete), with additional outside fireproofing.

> Tape isn't dead, but it's not worth it for small quantities

The cheapest LTO-6 drive on NewEgg is $1500, and Sony has the tapes for $18/TB. External hard drives are running about $35/TB. So you need ~90 TB for cost crossover on sheer data volume, not considering usability and reliability. So I would agree, with those kind of prices, you might want to *start* thinking about tape when you get to 100 TB, because 1 drive isn't very reliable. It might work for backup storage, since you can get by with a broken tape drive for however long your backup cycle is.

> That is completely self-referential.

Nope. A "bitcoin" unit is just an entry of 1.0000 in the transaction ledger known as the "block chain". The block chain is just a bunch of files listing every bitcoin transaction ever. My copy is 36 GB at the moment.

The Bitcoin Network is what makes it possible to write new transactions into the ledger in a secure way. Secure means nobody can rewrite old entries in the ledger, and everybody can verify the contents are correct. Only the person with the private cryptographic key to an address can send the balance in that address to someone else. Without the network, the ledger could not be updated, making the balances recorded there useless.

> First, a Bitcoin in of itself has no real-world value.

Neither does a UPS shipping label. It's the network of trucks and distribution centers that give the shipping token (the label) value. They are what enable moving a package from one place to another. That's a useful service, and hence people are willing to pay for the label.

Similarly, a bitcoin is merely an entry of 1.0000 units in a big distributed ledger (the block chain). It's the network of relay nodes and miners that give the bitcoin token value. They are what enable moving monetary value from one place to another. That's a useful service, and hence people are willing to pay for the tokens. Other parts of the ecosystem add more usefulness, and thus more value. Websites, wallet software, custom hardware, smartphone apps, exchanges, merchants who accept bitcoin, etc.

The transaction protocol also includes a scripting language, so you can make your money programmable. How useful is that? People have only touched the surface of what you can do with that capability.

Except the Bitcoin Network already runs at 324 Petahash/second, and each hash computation requires many floating point operations - 128 rounds of applying a complex hash function on several hundred bytes of data. Aurora competing for bitcoins won't make a significant difference in the network hash rate, it is too puny. The network already runs at ~1 million petaflops by dint of custom designed mining chips that perform the necessary calculations in hardware, massively parallel in each chip. Then you aggregate server rooms full of these chips into a mining farm.

You will need thrusters to position the big satellite in the first place (or the parts if it is assembled on site), and to counteract the Moon and Sun's tidal forces. If you can handle those, light pressure will be a small issue.

> I think a large problem is going to be space debris -

Nope. If you can build giant solar arrays in GEO, you can build small ones and attach ion thrusters to them. See the Dawn mission at Ceres and the Asteroid Redirect Mission NASA is proposing for examples. These space tugs can putter around and collect loose space debris. That however does not eliminate natural meteoroids. So your power satellite will need a maintenance program, or just accept a small amount of degradation as stuff hits it.

Solar arrays are thin, so most debris will just punch a small hole.

> The biggest unknown is the microwave link to send power to Earth.

We actually have tons of data about this, from all the GEO communications satellites, and rain fade that happens sometimes.

> The next-biggest unknown is availability of construction materials.

I was one of the people who worked on this issue while at Boeing. We found that 98% of the materials for a solar power satellite can be obtained from the Moon. A higher percentage are available if you use the Moon + Near Earth Asteroids. We didn't do the numbers for the NEO case back in the 1980's, since we had only discovered ~150 back then vs 12,500 today, and ion thrusters were not fully developed until about the year 2000. A modern study would account for both sources of materials.

We solved that problem early in the Solar Power Satellite studies at Boeing. The microwave transmitter in orbit is a phased array. The reference signal to adjust the phase is a transmitter in the center of the rectenna on the ground, powered by the rectenna. If the beam wanders off target, no reference signal, and the beam is no longer focused.

Hi Maury,

* Spectrolab rates their space solar panels for 20 years at GEO: http://www.spectrolab.com/Data.... Since they don't need to withstand weather, they can be much lighter than ground-mounted panels. 13 W/kg for a typical ground panel (not counting mounting and tracker) vs 177 W/kg for the space ones. That has implications for the energy payback time if you manufacture the panels in space.

* Your comparison of operating hours neglects that in space you have 36% higher insolation, because there is no atmospheric absorption. Therefore it takes fewer cells to produce the same output. Also the Nevada desert is an excellent location on Earth. The average location on Earth gets considerably worse hours of sunlight. Since we can't transmit power all over the Earth, cherry-picking a good location is unfair.