Disruptive Innovation with Open (Source) Hardware
Not really that different to open source software, or is it?
A good definition of what Open Hardware is, can be found on the website of ladyada:
“An open hardware license is basically an agreement by the author(s) of the work that allows other people to use that work for free, without paying royalties, licensing fees, etc. as long as certain constraints are followed. The most common constraints are "attribution requirements," for example requiring anybody who uses the work to place the name of the original author(s) prominentlyon the final project. Another common constraint is "share-alike" that means that any derivative work must be released under a similar license.”
At first sight, there is no real difference compared to open source software.
Advantages of open hardware?
As with any open innovation initiative, it will reduce R&D costs and increase community driven innovation. The software industry has proven that this works for operating systems to webservers to open source office suites.
Being open makes collaboration easier which means that there are peer reviews to keep quality high and to expose design flaws early on. This is especially important for hardware design where it can take a long time to test.
Collaboration means also that more people are working on it and this will also decrease development time and speed up feedback (taking into account the mythical man month of course). Ideally an open design -for example from SparkFun- to start from is available and will just need some tweaking to create new piece of hardware.
So far, the advantages are very similar to Open Source Software advantages.
Difficulties with open hardware?
The biggest difference with software, are the multiple hardware layers that can be licensed. There is the mechanical layer, schematic layer, layout layer, part lists, FPGA code, firmware code and software on top of the firmware. To be truly open, each part has to be open.
Hardware designs are also much harder to test compared to software. It is an expensive and slow process, but 3D printing is making that easier. Voltera is working on a printer to make prototyping easier and to even print a small batch for production. If you just want a simple solution to play with or test simple things, you can even use an existing inkjet printer with AgIC circuit printer.
Another difficulty is version control or -even better- change management as done with GIT with full branching and merging support for changes for each layer. This is quite hard to do since most layers as described above are linked to each other and must follow the same flow.
The integration tests will also be much harder to do and it is not easy (firmware/software) or even impossible (actual hardware) to upgrade after a release and this could lead to forking the whole design.
Not at the very least, there are no clear proven licensing strategies yet. For the pure software part and the documentation, respectively Open Source Software licenses or Creative Commons could be used, but for the integrated solution as a whole, organizations are still investigating how to deal with this.
So where does the money come from?
Open Source Software developers need to eat and support a family too, and so do Open Source Hardware developers.
The most obvious part to make money is by selling the hardware itself, but another option is to sell an end product or a result of using the hardware. A good example is to sell 3D prints instead of the 3D printer itself which is a piece of Open Hardware. Another very concrete example is the Open Compute Project, an initiative of Facebook to create efficient server, storage and data center hardware designs for scalable computing which Facebook uses also in its datacenters.
A third option is to sell closed addons and on demand development of extra features and addons. This happens also a lot with Open Source Software where a commercial company offers for example a better user interface or more/better integrations with other products.
A last set of opportunities I can think of are more traditional in the sense of teaching, consulting, coaching, support, but also renting and re-selling are good candidates to generate income.
Is the market large enough to invest in this? If we may believe Gartner, there will 4.9 billion connected ‘things’ in use in 2015 and 20 billion in 2020.
It is clear that what Open Source Software did to the software business in the 90s, is happening right now to the hardware business through Open (Source) Hardware. It is a little bit more complex, but it is just a matter of time before it will be common good.