Donal Reddington on mass customization, crowdsourcing and digital manufacturing

Can OSCar move from computer to garage? (Car Trouble, Part 2)

Recently I wrote about the difficulties that the U.S. auto industry finds itself in, due to its over-reliance on the build for inventory business model, as well as dependency on SUV sales for profitability.

At the end of that posting, I asked the question as to what might happen if there was a disruptive change to the business environment in the auto market. While there is no evidence that I have seen to suggest such a change will take place anytime soon, it is worthwhile to look at some projects that could someday ‘move the game on’.

What would happen if, for example, another concept were to be imported from the IT sector, namely Open Source?

A project that is exploring this possibility is OSCar (Open Source Car). OSCar began in 1999 with goal was of building a car “without an engineering centre, without a boss, without money, and without borders…but with the help of the collective creativity of the internet community”.

The long term prospects of the open source model depend on many elements, such as the legislative environment for vehicles in different countries, and the financial cost of constructing a car compared with the cost of simply buying one. In this posting, however, I am limiting the scope of the discussion to the more basic question of how an open source automotive design might move from the computers of designers to the garages of builders.

It was envisaged that the OSCar project would involve a development timeframe of about three years to get from initial project definition, through prototyping to eventual production of cars by individuals, based on the agreed design. However, the project has not progressed beyond design stages to date. Project founder Markus Merz states that it is just a hobby for those participating – the targets are not ‘business goals’ that must be met.

However, there may be a more fundamental reason which explains the slow progress of OSCar: people who work in IT or are interested in Open Source development issues are unlikely to be also skilled in metalworking or car mechanics. While it is a generalisation to say this, it is a fair assumption that someone who has studied programming or CAD will not have spent their evenings in their garage using a welder. Therefore, without the active participation of people who are familiar with the necessary construction skills, there is an increased risk that OSCar will not become a three dimensional reality.

This type of disconnection between the distinct areas design and production can adversely influence the progress of both. The question of separating design from production, and how the absence of production capacity can damage design ability (i.e. damage innovation) was discussed by Michael Bauwens on the P2P Foundation blog in January. He quoted from a New York Times article, that looked at the dangers to innovation posed by reductions production capability, but in the context of offshoring.

The NY Times article asked:

“Over the long run, can invention and design be separated from production? That question is rarely asked today. The debate instead centers on the loss of well-paying factory jobs and on the swelling trade deficit in manufactured goods. When the linkage does come up, the answer is surprisingly affirmative: Yes, invention and production are intertwined.”

It goes on to quote Stephen S. Cohen, co-director of the Berkeley Roundtable on the International Economy:

“In order to innovate in what you make, you have to be pretty good at making it – and we (the U.S. economy) are losing that ability.”

Michael Bauwens agrees, and sees an example of this in open source software:

“There is a certain truth to that. For example, one of the reasons for the success of free software in projects such as Linux, is clearly that design and production are one and the same in software, and that there is no hard frontier between developers and users, so that user insights can be integrated in the production process.

Would that mean that a project like the OSCAR open source car is doomed?

The answer would seem to be obvious: to succeed, its needs some kind of embodiment with production, which would suggest that the production process itself would need to be open as well, and that is much more difficult to achieve, as long as physical production is proprietary. Unless an enlightened physical producer opens up to the designer community.”

It is somewhat unlikely that a large scale physical producer would open up in this way. Why would they want to? Embracing open source design would be seen as ‘diluting their brand value’.

There may be an alternative approach to seeking the goodwill of a large scale manufacturer. There is an existing community numbering in the tens of thousands who could provide the production capacity required to bring a project like OSCar from concept to reality. They are the hot rodders.

In addition, there is an existing auto design which has become a de-facto open standard, providing evidence in support of the proposal that open source can be successfully applied in the automotive sector. Many different smaller manufacturers offer products and components based around the same basic architecture. This de-facto open standard car is not some present day design project. Enter the Ford Model B.

This car was produced by Ford between 1932 and 1934. For various reasons, it is the single most popular design among the hot rod community. Tens of thousands of hot rod cars based on the Model B have been built around the world over the last fifty years or so. There are many different companies that produce components that are compatible with the overall architecture of the B, such as Total Cost Involved, Roadster Shop, and The Rod Factory, who make reproduction chassis, Heidts who offer suspension components to fit those chassis, and many others who offer reproduction bodies in fibreglass and steel. These are used by both amateur and professional hot rod builders to construct hot rods based on the architecture of the Model B.

The point of all this is that the design parameters of the Model B are widely known, and there are many independent producers of the components. However, in many cases, these components are compatible with each other, because they are based on a ‘standard’ architecture, namely the original dimensions and design of the Model B.

So, if the OSCar community could arrive at a ‘Beta’ design for the Open Source Car, and then engage with amateur and professional hot rodders by posting design details on hot rod community websites, it might generate curiosity among some builders, who may be tempted to try out this new idea. The possibilities of this would be increased if the published architecture is sufficiently robust (in the physical sense) to enable high-performance engines and drivetrains to be used. (Open Source transport does not have to be synonymous with ‘eco-friendly’ or ‘slow’.)

Another automotive community that could contribute to the development of OSCar are those who build component cars (sometimes called ‘kit cars’). Component car builders are likely to be culturally closer to the OSCar community than the hot rodders. However, they tend to prefer assembling from a pre-existing set of components, rather than fabricating from scratch.

By providing effective feedback mechanisms for those who would take on the beta design and construct examples, improvements could be published for approval by both designers and constructors. These approved changes could then be incorporated into the later iterations of the design, before a finalised ‘Version 1.0′ is published.

I haven’t engaged with the OSCar community in preparing this post – it has taken a while to transfer these thoughts to writing and I’m not sure I could have explained them clearly in advance of writing the entire post. However, I welcome comment from anyone who is interested.

Share this item on these services:
  • digg
  • Furl
  • NewsVine
  • RawSugar
  • YahooMyWeb

Leave a Reply

You must be logged in to post a comment.