In the future will everything be open source?

January 27, 2007 11:52 AM

Can you build an open source car? Or a three-dimensional printer? Or a new kind of handheld computer? Can open source ideas thrive in the physical world? Or is there something fundementally different? Are we seeing a new type of production or just a filip for hobbyists and dreamers?
posted by MrMerlot (26 comments total)

Previously on the open source car.
posted by quin at 12:37 PM on January 27, 2007

The OpenBook project is a neat idea, but like all open source hardware ideas, the problem becomes: how do you get it manufactured?

Open source software thrives because it's easily transportable. There are no manufactures that need be involved. The developer can also be the distributer to the rest of the world. When hardware comes into the picture, you suddenly have to contend with fabrication, distribution, packaging, etc. And at every step of the way you will probably be dealing with people who aren't seeing this as a 'community project' but a paycheck for something they are physically building.

I would love to see OS hardware overcome this, but it's going to be a difficult trick to pull off.
posted by quin at 12:49 PM on January 27, 2007

China, my friend, China.
posted by TwelveTwo at 12:52 PM on January 27, 2007

All we have to do is get some open-source 3d printers manufactured, and then everything else is golden.
posted by adoarns at 1:05 PM on January 27, 2007 [1 favorite]

(from the Tommelise site)
While the part replicated, a drive rod gripper, was small (4.8 cubic centimetres), its creation represents a major milestone in the technology drive to create self-replicating machines capable of creating a variety of useful products.

I'm not trying to be a stick in the mud, but someone please, PLEASE tell me why we need self-replicating machines on planet Earth.
posted by Laugh_track at 2:00 PM on January 27, 2007

I don't go in for gadgets really, but I'd probably buy an OpenBook if it ever materialized.

As far as open source methods go, some things -- like the Free Beer project -- don't really seem to have a point. Recipes abound, and you can modify them to your heart's content. And you'll play hell trying to actually duplicate the beer, since water in different cities has different mineral levels, etc. Slapping an "open source" label on a beer recipe just seems cheesy. Cheap publicity for Vores Øl, I guess.

(Also, guarana beans? I'm by no means a purist when it comes to beer, but that's pretty nasty. Feh.)
posted by cog_nate at 2:02 PM on January 27, 2007

I'm not trying to be a stick in the mud, but someone please, PLEASE tell me why we need self-replicating machines on planet Earth.

Because of economies of scale. If you create 10 machines and tell them to replicate 10 times, after enough generations you have machines capable of creating something new (other than themselves) at a much lower overall cost.

People are especially interested in this for machines on the microscopic and less scale. Think -- nanotubes, buckyballs, etc.
posted by WetherMan at 2:05 PM on January 27, 2007

See also fab@home (open rapid prototyping machine), here's a nice video demo.
posted by MetaMonkey at 2:06 PM on January 27, 2007

Well, Forrest Higgs's machine Tommelise (in the second link) is GPL, and the fellow working on it making it from wood using hand tools. Homebuilt 3D printers are what our group is calling RepStraps, in that they can be used as a bootstrap machine to build a RepRap. A RepRap is a 3D printer made using another 3D printer, rather than conventional machine tools or hand tools. It will look like the machine in this photo or this rendering.

This kind of 3D printer, (and the fab@home machine) is called a Fused Deposition Modeling machine. Conceptually, it's a hot glue gun or motorized syringe mounted on a 3-axis positioning system.

(The white plastic parts in the photo are made using a ~USD$20K conventional 3D printer. For RepRap, we'll make the 1.0 announcement when we've made a RepRap using the printed parts from another RepRap. The build cost for that will be less than USD$500. (It's not a perfect Von Neumann Machine, you still need to go buy steel rods and stepper motors, but we're hoping it will change things.
posted by sebastienbailard at 2:12 PM on January 27, 2007

I'm not trying to be a stick in the mud, but someone please, PLEASE tell me why we need self-replicating machines on planet Earth.

I think the basic idea is that technology requires a great deal of groundwork to be laid. Each generation of technology is built on the generation before. We're a very long way up the technology pyramid, and if any piece of it crumbles, the whole structure can collapse.

The fundamental idea behind these manufactoria is that they should be able to A) build all the parts necessary to create another one of themselves (even if they can't actually self-assemble) and B) build anything else at the same or lower technology level. This is an exceedingly hard problem, but if we can solve it, it will mean that civilization will have a much harder time completely failing from major catastrophe. A machine that sophisticated would allow the reconstruction of our technological base from, essentially, nothing.

That said, the overall idea is so far removed from what we can do now as to effectively be magic. These early attempts are crude... they require things like refined metals as inputs. A proper manufactorium would be able to start with rocks, scrap, and absolutely idiot operators, and still produce finished goods. It would be technology so advanced, in other words, that it would look to us much like jet fighters would to cavemen. :)

Even a small-scale version of this would be useful, though. A tool to make tools, even if it requires refined raw materials, would be handy to have around the house if you like to tinker.
posted by Malor at 2:18 PM on January 27, 2007

Well, humanity's a long way from a molecular assembler, which I suppose is the platonic ideal for a self-replicating 3D printer. On the other hand, that means wwe don't have to worry about grey goo just yet.
posted by sebastienbailard at 2:54 PM on January 27, 2007

Another one you missed, which I gather is still a work in progress, is the open source business (also here).
posted by scalefree at 3:36 PM on January 27, 2007

I'm not trying to be a stick in the mud, but someone please, PLEASE tell me why we need self-replicating machines on planet Earth.

So that people can afford one.

Then, rather than shipping mass-produced goods around the world, goods can be made locally on an as-needed bases, and they can be customised to particular needs, instead of the one-size-fits-all mass manufacturing we have today.

There was a time when if you wanted a table for a room, that table was made specially for that room. Fast forward a few hundred years to today and while more people can afford tables (many more people), they still can't afford a table built to their unique needs. Fast forward a hundred more years and hopefully you get the best of both worlds - products that are both affordable and customised to the need.
posted by -harlequin- at 4:26 PM on January 27, 2007

I think the Onion mentioned this.

Sorry! Had to! :-)

Actually, I envy the guy having so much space in his house that he can do that. I have a sentimental attachment to old computers, but after a while, their state of I-never-use-them and I-will-never-use-them-in-the-future-either starts to grate and I feel better getting rid of them.
posted by -harlequin- at 4:36 PM on January 27, 2007

Gah, wrong thread. Ooops.
posted by -harlequin- at 4:36 PM on January 27, 2007

Also, there are special-purpose items, like braille typewriters or openprosthetics.org which are expensive to manufacture and expensive to make replacement parts for right now because they're made in small quantities, by conventional machine tools and injection molding equipment, etc, etc. The manufacturers don't experience economies of scale, and there's not lots of competing factories making them, so they're not cheap.

Cheap and widespread 3D printers will hopefully change the economics for that sort of thing.

I think aside from specialized applications, the bulk of early adopters will be people doing hobby robotics, mabye. It's hard to say.
posted by sebastienbailard at 4:42 PM on January 27, 2007

Open-source projects for the musically-inclined:
x0xb0x, the open-source tb-303.
eDrum, the open-source electronic drum controller.
You can keep your grey-goo-manufacturing nanited. I'll be over here playing with my noisy toys.
posted by lekvar at 5:05 PM on January 27, 2007

Art-wise, there's also the Creative Commons Arduino board, for doing interactive electronic art stuff. The Arduino board is a smaller cheaper ($32) version of the Wiring I/O board
($80)

These are both microcontroller cards, USB programmable, with lots of pin outs, programmed using the programming language Wiring, which is really easy to get in to.

This is great for people who want to make a cyborg teddy bear or something, and they don't feel like designing their control circuitry themselves.

Wiring is an outgrowth of the programming language Processing. Processing is a stripped-down version of Java intended to for algorithmic art. (Note that you can access all of Java through processing, but processing provides a very simple IDE for non-programmers to jump right in.)
posted by sebastienbailard at 6:48 PM on January 27, 2007

I have different requirements for a car!

Asking for an open source car is like asking for an open source laws of physics - not going to happed.
posted by Samuel Farrow at 9:04 PM on January 27, 2007

The fundamental idea behind these manufactoria is that they should be able to A) build all the parts necessary to create another one of themselves (even if they can't actually self-assemble) and B) build anything else at the same or lower technology level. This is an exceedingly hard problem, but if we can solve it, it will mean that civilization will have a much harder time completely failing from major catastrophe.

Thanks for the reply, Malor. But how does civilization fail? For that matter, how does it succeed?
I love technology -- I use it all the time. I'm using it right now. But you have to admit that arguments for the spread of technology on the basis of technology's self-evident merit are sketchy. Nothing about increased technology guarantees a more peaceful, habitable world. Think of how much easier it would be to manufacture deadly weapons or instruments of slavery with the "manufactorium".

Fast forward a hundred more years and hopefully you get the best of both worlds - products that are both affordable and customised to the need.

This answer is a little harder to counter... but I still think that the risks to human sovereignty outweigh the potential gains in comfortable furniture. Just a base reaction, I guess. Nice response though.
posted by Laugh_track at 9:26 PM on January 27, 2007

The problem with the open source car is certifying it for emissions and OBD-2 compliance.
posted by rfs at 10:03 PM on January 27, 2007

Frankly, I'm more interested about the current and near term social and economic consequences of open source hardware than hypothetical threats to society posed by nanotech. This is because we've (not necessarily in this forum, but as a society, and as geeks) already discussed and thought about grey goo, along with golden goo, black goo, red goo, khaki goo, blue goo, pink goo, and green goo.

I mean, I don't mind going on about clanking replicators, drexler engines, and hard singularities, but I don't think any new insights will come out of us showing off our l33t sci-fi chops.

We haven't really discussed what the consequences are if cottage industry comes back into being in the first world, even as manufacturing has fled it. What does it mean if you can print out most of the components for a sewing machine, rather than buying it at Walmart? Is legos still going to be around 5 years from now? Are sherrif's deputies going kick down your door with a restraining order if your kid prints out mickey mouse figurines? Those are questions we don't know the answers to.
posted by sebastienbailard at 11:04 PM on January 27, 2007

See also, The RepRap project.
posted by Jerub at 1:09 AM on January 28, 2007

Jerub, it was mentioned upthread and I'm actually one of the developers. I was wondering what other people think of RepRap in specific and Open Source Hardware in general.

A few questions:

Do people think cottage industry is going to come back? How is Open Source Hardware going to change things, if it will really?

What would you make with a 3D printer? parallax7d asked some time back about making his own computer mouse, and there was Riovanes's topic about making furniture from 3D scribbled input using stereolithography (crossed laser beams, tank of goop). (There was also the Polecat "laser printed" military drone. but I don't think Lockheed Martin is going to open-source it for all those hobbyists who want to spy on/shoot missiles at their neighbors.
posted by sebastienbailard at 3:23 AM on January 28, 2007

I have a blind coworker, he loves the idea of just being able to print things so he can feel them and see what they 'look' like :)

I see projects like the reprap project being incredibly useful in education. You can print anything from toys to electronics projects to educational aids.

Need another 3d printer because you've got another concurrent science class? Print one.

Have a promising young student who would love to have a 3d printer for home? Let him print one after school. :)

On the other hand, Cory Doctorow wrote Printcrime</a, the worst case of 3d printers being economically viable to run.
posted by Jerub at 5:15 AM on January 28, 2007

I'm not trying to be a stick in the mud, but someone please, PLEASE tell me why we need self-replicating machines on planet Earth.

I'm assuming you mean artificial machines. There are, of course, a few self-replicating machines I could do without, notably Iranian and North Korean examples, some of the posher American ones, too—but all in all I think they serve their purpose.
posted by adoarns at 7:09 AM on January 28, 2007