Ben, I think the conventional bicycle - and by now a good number of unconventional ones - represents technology that has been thoroughly vernacularized over more than a century. It is pretty much open-source already!

In fact, the mix-and-match nature of a bicycle is something worth studying for other projects. Designers are too blindly wedded to the idiom of the modern product, which has been developed expressly around the closure (indeed, the enclosure) of sources, and therefore design products around sets of complex, exclusive, over-critical component interfaces.

Take the example of car headlights. Before c.1932 the norm was for headlights to be mounted either to a horizontal bar or to pedestals on the frame horns. This allowed a very wide variety of headlight configurations to be applied to any given design. The component designer did not have to deal with any but the simplest mounting detail, and thus a wealth of design responses is stimulated. This is a quality that survives in the basic bicycle. One might say that it is stigmergically open.

For a long time after that the norm was a relatively simple circular, flat, annular-disc interface. This was not as stigmergically open as the prior arrangement, mainly due to difficulties in achieving an aesthetically happy arrangement without compound-curved sheetmetal directly adjacent to the flat headlight-mounting area; but it was a lot more open than the current norm.

A former girlfriend of mine taught me all about good reasons and real reasons. Here the good reason is aerodynamic efficiency, more often than not. The real reasons are aesthetic, on the face of it, that is to say marketing; but really have to do with the closure of sources and the restriction of an industry to technologies of manufacture that are accessible only to a privileged few.

A really ecological open-source way to build bicycles could be to use bamboo poles for the frame, and fit them into adjustable joints that could be produced by additive fabrication.

maybe the joints could be produced by additive fabrication and then reinforced by hemp which i've seen many pictures of. Another challenge would be building wheels. The wheels consist of a hob, spokes, nipples, rims, rim tape, tube, and tire. There are a lot of options. An extremely simple bicycle such as fixed gear bicycle consists of a frame, bottom bracket, headset, fork, stem, handlebars, seat post, saddle, wheels, cog, crank arms, a chain ring, and pedals. Hand brakes are optional since the way a fixed gear works allows you to break using your legs. My interest would be to create a way to document and distribute the information needed to build each of these parts to make a simple and functional bicycle from raw materials. I also do not have too much experience so I will need to learn a lot about each of the components in order to do this.

Ned: Of course one point of using bamboo in construction is that it should be grown locally, and so provides a renewable low-energy building material.
As for "the amount of engineering input required to convert the material to a usable form is not such to defeat the object", I'm really not sure what you're talking about when you say 'defeat the object', nor later when you say 'the issues'.
Are you perhaps referring to the need to adjust a joint to fit various different diameters in bamboo poles? If you have experience of using bamboo in architecture, is there any particular way you find is best for securing the ends of varying poles? I was thinking of something like either hose clamps, set screws or lashings, but I think a set screw would eventually tear out of the bamboo with vibration.

Ben: I'm not sure what you would stick hemp reinforcement fibre on with, but there's no reason why you can't for instance print a thin-walled model of a bicycle joint in PLA and use it to sand cast the joint that you'll actually use out of recycled aluminium or brass. I don't think direct metal sintering will be an option in most hackerspaces for the next couple of years though.
To produce wheels, you either need access to equipment to form sheet metal with, i.e. bend a strip into a hoop and bend its edges up as a rim for a tube & tyre to be seated in, then the wheel can possibly be spot-welded together, or you need to steam a strip of wood in order to bend it into a similar form and probably fix it with adhesive. Spokes can be practically anything, but a hub is a difficult bit, especially a ratchet-freewheel type that allows you to apply force in one direction. I would be inclined to buy in or salvage a stock component, unless its users don't mind having their legs constantly in motion while going downhill due to not having that mechanism...
It takes a fair bit of land to produce enough latex for tyres, but making a mould for them shouldn't be terribly complicated. Inner tubes would probably be more difficult since they must be much thinner.

(Also keep in mind that dandelions are good for a lot more than rubber, if viewed organically.)

Hi Ben

joeykork, I think I see where you have misunderstood me. When I say, "Designers are too blindly wedded to the idiom of the modern product ..." I refer to product designers in general and those looking at open-source or other alternatives in particular, and not bicycle designers. Indeed I recommend a thorough study of bicycle manufacture instead of the prevalent idiom in the field in question. That is, if the component-interface regime on an open-source car, for instance, works like that which has traditionally held for bicycles, rather than the (rapidly shifting) current norm for automobiles, we shall have gained.

The bicycle industry is one relatively free from "nefarious closed ecosystems"; for which reason I recommend it as an example of how other industries might work.