Author Archives: Matt Maier

Applying Open Source to Rocket Landings

Blue Origin landed their booster back on the ground after lifting a capsule into space (or mostly into space, whatever). This is an awesome development. We need rocket landings and not just rocket launches to make them as reusable as airplanes to bring down the cost of spaceflight.

SpaceX and Blue Origin Rocket Landings

Of course, Jeff Bezos talked about what Blue Origin achieved and, of course, Elon Musk of SpaceX and other fame felt the need to throw a little shade and talk about how SpaceX is better. Billionaires will be billionaires.

I expect we can get this kind of development out of the way faster, and make everything about space cheaper, if we adopt an open source approach to the technology. Leaving aside concerns about export controls for the moment (which pretty much would put them in jail if they shared anything with people outside the U.S.), it’s just cheaper to solve problems if you already know what’s worked for someone else.

Open source was invented so that companies could safely use public code in their for-profit endeavors. People can use it too. Aliens could use it if they wanted. Anyone can use it, but the “people” who are happiest are the corporate lawyers because they don’t have to worry about their company getting sued. It’s an important distinction because sharing is an old idea, and open source is a lot like sharing, but open source ain’t sharing. Open source is a legal standard, not a moral or ethical standard. When individuals want to share they rarely bother applying a license to their work.

ohw_logo-golden_orb

There shouldn’t be any doubts about guys like Bezos and Musk knowing what open source is. Bezos clearly knows what open source is and intends to use it without commenting or contributing. Musk actually invoked open source when he allowed the world to use Tesla’s patents. However, in contrast Musk also said that SpaceX doesn’t patent anything because then China would know what they’re doing and copy them. Bezos tried to patent the idea of landing a rocket but Musk helped block the patent, arguing that tons of people already know how to do a rocket landing. So, if they know what open source is, why don’t they recognize its potential for creating cheaper solutions? I’ve got a theory on that, but first a bit of context.

There’s a bit of tension, then, between the perspectives of individuals and corporations. The corporations (SpaceX, Blue Origin, etc) only “share” when it’s safe and profitable. The masses share pretty freely, but only license what they share some of the time. So if we wanted to get lots of people throwing in on designs for reusable rockets we’d need to reconcile the difference between those perspectives.

Open source works best for infrastructure type technology; stuff that everyone needs. So we might be able to build up technology from the masses that’s used in many things and can also be used in spaceflight. A more fanciful possibility is using something like a Distributed Autonomous Corporation to organize the effort of the masses around actual spaceflight technology without a traditional corporation. We might even be able to eventually educate the people who run spaceflight corporations on the benefits of sharing development costs, but this is supposed to be a non-fiction piece.

There’s inherent issues with Open Source though when it comes to hardware. There’s a lot of activity in open source software because there aren’t really any barriers to building source code into object code, whereas things like an open source 3D printer (see RepRap for probably the first), you need to buy lots of materials that you might not otherwise have, and then, oh, you’ve got to build it. There’s not some machine that’s going to do it for you like software. There’s activity around many electronics projects, and a few mechanical projects, because turning the source into object isn’t too onerous. You can perhaps see how something like a rocket get’s hard to open source.

For something like rocket landings though, most of the systems are already in place. You already used the power of the rocket to go up, so slowing down can be done the same way (and is by Blue Origin and SpaceX). Just fire the rocket in the same direction, away from the ground, again. That’s super over simplified, but it’s to make a point. If Bezos and Musk (and NASA and ULA and maybe more) decided to start working together on the problem of landing, they already have 80% or more of the things they need. The last 20% is a bunch of hardware and software performing tricks for controlling the rocket on its way back down that they both sort-of have working, but it’s not 100%. Even if it has been proven to work after multiple orbital launches, there’d be more things to improve on. Like landing on a barge in the middle of the ocean where it’s moving around tons.

Falcon 9 Landing Stages

Landing the Falcon 9, via SpaceX

They could start open sourcing the landing systems so that they can do hard things like landing on a barge in the middle of the ocean with the likely super complex software SpaceX designed to take in all sorts of information from sensors all over the Falcon 9. It could then be worked on collaboratively with other organizations (again, NASA, ULA, etc). They could even sweeten the deal with a list of the sensors that they’re using with it, where they go on the rocket, etc. If they aren’t commercial, off the shelf (COTS) items, things like that would be great to share their design and manufacturing with the likes of Blue Origin under an open source license too. With that done, as Blue Origin uses that software and those sensors and integrates them with their rocket and whatever they’re using to “balance a rubber broomstick on your hand in the middle of a wind storm” (via SpaceX) that is rocket landings from space, they can suggest ways to improve it, or even contribute their own improvements! All of the other bits of the rocket like the engines don’t have to be open sourced as well.

What’s better, is now the rockets are cheaper for more reasons than one. Yes, they’re reusable. But there’s not nearly enough people doing the same thing at 2 or more different companies. Now they’re working together, and that means developing better landing stuff faster, whether that’s software controlling the rocket, or the hardware helping it stay upright and not turn into a giant fireball (see videos of attempts at rocket landings on a barge).