Sunday, April 13, 2008

What is open source hardware? Part 1

What is open source hardware really, and why is it so much harder to define than open source software? Recently, there seem to be many attempts to define Open Source hardware, but very little agreement, and certainly no official "expert" who can claim the answer. So where to start? Well, there appear to be two parts of this question:
  1. First, what is Open Source hardware?
  2. Second, what are the collaboration rules of open source hardware and how are they different from Open Source software?
Before we begin, we need to understand how Open Source hardware (OSH) is different than Open Source software (OSS):
  • OSH is tangible, OSS is intangible
  • Each OSH product must be manufactured, whereas OSS can be replicated infinitely
  • OSH requires $10k's of fixed investment (equipment) and $1k's of variable cost (pcb runs and parts), OSS requires $1k's of one-time investment (PC) and $10's of variable cost thereafter (e.g. internet connection)
  • OSH involves making physical things, and there is a cost scale barrier for most products; that is it costs a hardware hacker $500 to make a single $50 product from the ground up, but if he can find 100 other people, the product now costs $55
  • OSH requires specialized roles for circuitry, pcb, placement, and soldering design that require contracted services from companies that maintain complex machinery, whereas OSS is generally the same role easily accomplished by a single person start to finish

So what is Open Source hardware?
If we believe the list above, Open Source hardware is about manufacturing and building real stuff, is an order of magnitude more expensive than software, and requires lots of people and companies to pull off. Classical economics would say open source hardware is the production of tangible assets victimized by scale economies and belabored by high coordination and transaction costs. I would simply say it's more complex than Open Source software.

The complexities outlined above suggest a root problem; that the success or failure of Open Source hardware is linked closely with what a single individual can achieve within a reasonable personal budget. In part 2, I'll talk more about this; what seems to be the "Open Source Hardware productivity problem"... before Part 3, which will talk about the collaboration rules for Open Source hardware communities.

No comments: