22/06/10
It is just over a year since we introduced the 40 Fires Foundation to the world, when Riversimple unveiled its demonstrator vehicle in London. Here are some reflections on the journey we have been on.
We have set ourselves an incredibly ambitious task – to gather a community of people together to design collaboratively the world’s first truly open source car. It is also very much a journey of exploration – we have to keep trying things, seeing what works and what doesn’t.
It is clear that this project has huge potential. A large number of highly skilled people have been in touch with us, including experienced designers, engineers, entrepreneurs and students, and we believe this is only the tip of the iceberg, since we haven’t spent any effort on publicity, other than a couple of links from the Riversimple website.
At times in recent months I have felt impatient or frustrated when progress seems slow or when there’s little activity on the discussion forums. I have to remind myself that we are only at the beta stage - we haven’t properly launched 40 Fires yet. The fact is we have covered quite a distance since last year, even though there is a lot further to go.
We have a far better understanding of what an open source car actually means. For us, it means a collaborative design process and shared, freely available technical information. We can’t open source the whole car – many components like the tyres, the fuel cell and the ultracapacitors are proprietary and are likely to stay that way for the foreseeable future. But that’s not a problem. The secret to building the next generation of car is in the way the components are connected together – and that’s the knowledge we will share openly. And we will use open source components wherever possible, share the technology that we have developed and encourage the sharing of all new developments.
The main challenge for us is to tap into all the enthusiasm and energy that we have access to. This has proved harder than we expected. It needs time and energy to share knowledge and then to answer all the questions that arise. We can’t simply upload a few schematics on the site and sit back. Riversimple, who we depend on to provide the data, is still a small business, with limited resources. One of its main priorities at the moment is to raise funding. Once this has been achieved the development programme can kick off in earnest.
In the meantime our work is to lay foundations, building the base for the work to come. We spoke to Mark Shuttleworth (the founder of Ubuntu) recently and he suggested that there are four requirements for establishing a “digital commons”:
1. interchangeable formats for the content; 2. broad access to the tools; 3. an ability to work collaboratively; and 4. a supportive licensing framework.
Our work over the last year has been in these four areas and particularly the last three. We have been exploring the best tools to use for on-line collaborative design work and have had some very interesting and useful conversations with Dassault Systemes on this subject. It is clear that there is currently no freely available 3D design tool sophisticated enough to serve all our needs. Neither can we assume that all our contributors will have access to CATIA or some equivalent application. However, this needn’t stop us working in the short-term. A lot of sharing can be done using very basic and familiar tools, e-mail and discussion forums, using pdfs and scanned versions of marked-up drawings. The discussion forum we have been using, Nabble, is fine although there is probably something out there that works better to present all the various discussions that can be going on at one time. 3D images can be shared in IGS format and viewed on basic free 3D viewers or much more complex OpenSource 3D modelling softwares such as Blender. Many of our collaborators will have access to CATIA and other tools. And we will continue to pursue (although this is a medium or long-term project of its own) the nirvana of a low-cost or free sophisticated 3D application.
In terms of collaborative working, one thing that has become very apparent to us is that the car will not simply emerge from the community on its own. It needs a seed to be sown, which the community can water and fertilise. In the case of Wikipedia, the seed is the first line of the entry. In the case of an electric car the seed is an initial car design including schematics of the whole car, circuit diagrams, details of the motors and electrical systems. Although there is some data on the site, it is by no means complete. We will be working over the next few months to upload more but as I said before, our resources are limited so this won’t happen in a hurry.
Another aspect of collaborative working is having a shared understanding of the task to be achieved. One thing we will be working on over the next few months is defining the key design challenges and putting out a call for key experienced collaborators who can help us address them.
The fourth area mentioned by Mark Shuttleworth is licensing. This is about creating a moral and legal framework that holds the community together. We can’t hope to rely purely on goodwill from people and organisations to make the project thrive in the long term; they have to be encouraged to participate and to keep participating. Potential issues to solve include the free-loader problem (people won’t keep sharing if they think other people are taking unfair advantage of their generosity), dealing with patent trolls (organisations who are in the business of acquiring patents and then suing people in order to extort money from them) and working out how organisations can make money out of participating in 40 Fires (this is important if we want businesses to invest resources in contributing to the project).
For a long time we assumed that we would use a copyleft license, inspired by the GNU General Public license or GPL. However this is in general unattractive to commercial organisations who worry about the “viral” nature of such licenses (even though this fear is often illusory). More importantly, in a hardware world it is difficult if not impossible to enforce such a license, because of the very different nature of copyright versus patents. Instead we are moving towards adopting a so-called “academic license”, which will be less legally restrictive than a copyleft license, while encouraging a cultural norm in the community of sharing, which is vital to the health and success of our endeavours. An example of an academic license in the software world is the “BSD” license.
There has been a fair amount of discussion on the forums about this question and I will continue posting as we progress our thinking in this tricky area.
Looking ahead, we expect the following over the next six months or so: - a formal launch of 40 Fires; - schematics of the entire car to be uploaded on the site, meaning all the parts that Riversimple has designed itself, plus basic information about the proprietary components; - specific tasks for the community to work on, including development of the Vehicle System Controller which is likely to become the first fully open source component in the vehicle; and - partnerships with collaborators such as universities and commercial organisations.
This is tied in to a large extent with the progress of Riversimple’s fund-raising, which is due to be completed by the end of September.
For those of you who signed in the last year and are itching to get involved, please be patient. Your opportunity will surely come!
Patrick
05/11/09 Things are still relatively quiet on the discussion forum. We had hoped for more activity by this stage but I believe this is (at least in part) a function of how much we share. If we had uploaded the whole car by now I am sure we would have had many hundreds more views and loads of comments. The main reason we haven’t is that we have limited resources at our disposal, and lots of different tasks to get on with (in particular we are busy getting funding for our next stage). It is not simply a matter of uploading schematics – to be meaningful they need to be accompanied by a narrative explaining the design choices we have made.
There is also an awareness that we are learning fast and that by uploading a little at a time we can observe the response and adjust our approach accordingly.
Another reason is simply that a slight fear that if we share everything we have done so far, a rapacious and better financed competitor may come and take it all and leap ahead of us. We feel that we need to progress a bit further with our fund-raising before sharing everything.
Finally there’s the licensing question. Whilst the creative commons license we are currently using is fine for the designs of the technology demonstrator (which it is unlikely anyone will want to build), it will not do for the designs of the prototype which need proper protection. So we need to crack on with exploring this.
We will still continue to share. But not all at once. If the forum remains only with low–level activity in the meantime, that’s fine. I am confident that once we commit more time and resources to sharing, activity will pick up.
Patrick.
by Matt Asay.
There are signs everywhere that people are getting ready for OS hardware to take off. This is really good news for us - it is absolutely clear that we can't do this all on our own. Maybe someone else will do excellent work on a good collaboration platform for OS hardware, and we'll be able to share what we have learned (are learning).
Patrick.
No talk of energy efficiency or carbon emissions in the presentation, which is what we are really interested in. Still, it is good to see that there are companies seriously thinking about the future of transport.
Patrick
This feels like another mini-landmark as it is the first time someone not from the core team has been given authority to do something for the project. We will succeed to the extent that we are able to delegate successfully. This doesn't mean completely handing over control, without limit - it means a gradual handover of more and more of the day to day activities to other people and so gradually building a community.
It can be a bit scary to handover control but it is the most important thing to do.
Patrick.
When we first set up this wiki, one of the requirements was that visitors would not need to log in and that the whole log-in registration thing should be hidden. As a result (with a little bit of css jiggery-pokey) you can't see the normal "Log in", "Register", "Log out", "Preferences" bit that you would normally expect from wiki's built with the open source jspwiki engine that we use. However, if you were to hit one of our pages which was not designated for universal visibility you would be prompted to log in. Many people have obviously tried to register so that they can then log in and these registrations have then appeared to fail.
Let me come clean - we ignore all the registrations.
Even if you register with Nabble to post to the forums (which works) then you won't be able to log in to the wiki to make edits.
We are debating this and one day we might show the log-in features and start approving the registrations so that you can all join in. If you do think that you all should be able to do so then please do add to the debate in the forums!
One day we'll have a wonderful, open source, free or nearly free, collaborative, version controlling 2D/3D CAD system (any suggestions?) and a wiki that you can all contribute to but it's still early days.
Roland
We had our first workshop last week, which was enormously encouraging. Really good quality discussions, bags of enthusiasm from all the participants, and our first tangible experience of what is possible with this approach.
I have also been much encouraged by the response to the monthly e-mail we sent out on Friday. Several people have come forward to offer their time for specific work that needs doing. In particular a number of web developers have offered to help us improve the website. We'll have to work out a good way of harnessing their energy. Perhaps we'll hold a competition for the best.
We've also been looking into funding opportunities and Christian is pulling together an application to the UK Technology Strategy board. There are lots of opportunities out there for a project like ours, even in a recession.
Patrick.
Patrick
Patrick.
Patrick
The latter describes four stages of social arrangements in a collaborative project such as ours:
Sharing – e.g. Riversimple shares its ideas and its schematics and people look at them. Others start sharing their ideas too. Cooperation – People start commenting on the schematics and suggesting changes or alternatives, alternative components and the like. Collaboration – together we produce new designs for a production prototype. Collectivism – describes the development of a fairly tight group at the core of the project who review the work of the community and make most of the major final decisions (like the 1500 editors of Wikipedia who are responsible for the majority of the editing).
It is going to be interesting to see if this description matches our journey.
Patrick.
I will start with a fairly fundamental question - why go down the hydrogen fuel cell route at all, when there’s been so much excitement about rechargeable battery-driven electric cars? Aren’t they now pretty well set to inherit the role of the internal combustion engine (ICE) in the post-petroleum era?
There’s a simple answer to this, at the level of the vehicle, and a more general explanation.
The simple answer is ‘horses for courses’. If we want a sustainable transport system, efficiency is the key metric, and we cannot afford the luxury of a single solution that is as amazingly versatile as the ICE has been; compromised efficiency is unavoidably the price of such versatility. We believe that different types of powertrain will fill different niches, the key criterion by which powertrains are selected being the ‘installed range’, the range for which the vehicle is designed. Batteries will be more efficient over short distances but hydrogen fuel cells will emerge as the most efficient and flexible choice for inter-urban vehicles.
Why? Essentially, because the efficiency of a vehicle is very tightly coupled to its weight. There is a trade-off between the energy efficiency of the power source and the weight penalty incurred in carrying the appropriate fuel around on board the vehicle. This leads to different optimal solutions depending on the range of the vehicle. Batteries are highly efficient - say 70% or more for the round trip, energy in and out - but heavy. Converting petrol to rotation in an ICE is much less efficient - 20-25% - but the fuel tank and the liquid in it are relatively light; an easy route to an effective solution, as long as efficiency doesn’t matter. Hydrogen fuel cells come somewhere in between; efficiency might be say 50%, and the tank or other onboard storage solution adds a small weight penalty, but the weight of the gas itself is negligible.
You can extend (but not double) the range of a battery vehicle by doubling up the battery pack. Each successive battery gives a diminishing return, and you eventually reach a point where there is no gain in range from adding another; the line on a graph plotting fuel capacity against range reaches what is known as a vertical asymptote. This will happen, eventually, for any particular architecture and powertrain combination – the key difference being that it comes at around 500 miles for a typical lead acid battery concept, and 1,500 miles for an equivalent lithium-ion vehicle, but not until something in excess of 5,000 miles for fuel cells (and perhaps 10,000 miles for an ICE, making it pretty much academic). Moreover, at a range anywhere beyond about 10% of this asymptote, the vehicle's efficiency is severely compromised. My estimate is that a purpose-designed fuel cell vehicle becomes more efficient than a typical lead-acid battery electric vehicle at an installed range of about 20 miles; with Li-ion, this crossover is at about 100 miles.
The more general point relates to the whole system perspective, which is crucial for making overall sense of relating energy demand to sustainable supply. In whole system design, elements are not compared with their alternatives at an isolated level; the comparison must be made by evaluating the impact of choices at the whole system level.
Over the last century, our transport infrastructure and technology has become ever more inextricably linked with one energy source – petrol. That era is ending. Electricity, as an energy carrier, rather than a primary fuel, decouples energy sources from transport energy demand, as it ceases to be reliant on any specific source. However, to optimise access to all the renewable energy resources available, we should be wary of a pathway where all energy, in future including transport energy, has to be supplied as electricity. Electricity is, after all, the highest grade of energy there is, in terms of its versatility; we should value it as a premium product, and allocate it accordingly.
That’s partly why hydrogen has such potential. We must understand that hydrogen too is an energy carrier, an equivalent to electricity, rather than compare it with petrol. And a hydrogen grid, alongside the electricity grid, could tap a range of sources, such as biomass, algae and sewage for example. At the same time, although it would be inappropriate to generate all the hydrogen we need from electricity, it could mop up, store and distribute some of the off-peak output from intermittent renewable sources, such as windpower. This would deliver greater marginal benefit than any other use of these intermittent excess sources.
As these two vectors are eminently fungible, using them both in parallel allows us much greater flexibility in meeting specific demands; in combination, they can be more powerful than either can be on their own. In aggregate, by allowing for the most efficient and effective matching of energy sources to meet our many and varied needs, this allows us to wring much greater utility out of the renewable energy streams available to us. Coming back to Whole System Design, we cannot treat transport and energy policy independently and at a whole system level, the argument for hydrogen as an energy carrier in parallel with electricity is much more powerful than that for hydrogen for transport alone.
It goes without saying that a hydrogen grid is a major infrastructure investment. But, come what may, we’ll need to face up to the need for new infrastructure, even for battery cars. And why put big money into interim solutions? We must ensure that all investments lead us closer to our goal of a sustainable transport system, rather than down a blind alley. Dead ends are bad news, for two reasons; a) we don’t get to where we want to go and b) we have to write off the investment that has not got us there! Even while hydrogen is still coming from natural gas - ‘brown hydrogen’, so to speak, rather than the ‘green hydrogen’ sourced in future from renewables - the cleanest use of this natural gas is as hydrogen in a fuel cell. Using hydrogen from natural gas, our urban car has “well to wheel” carbon emissions of about 30 g/km, only a quarter that of today’s lowest emitting car, the Polo Blue Motion, and half that of the electric G-Wiz if you assume that G-Wiz’s electricity is also generated from natural gas. However, in the light of the urgency to rein in carbon emissions, we must not allow this to become an excuse to delay the transition to green hydrogen.
The eventual switch from brown to green hydrogen can then be made incrementally and without any further investment either in vehicle technology or fuel distribution infrastructure, and without the end-user even noticing; we can make this transition at any pace we can achieve. Furthermore, every region, anywhere in the world, can use the energy mix most appropriate to its resources, whilst benefitting from shared technology and standards. This buys us incredible flexibility to achieve the transition to sustainable transport – and do we need flexibility!
Hydrogen is much more powerful as a fuel in conjunction with the electricity grid. It will never be as dominant as petrol, or fuel cells as dominant as combustion engines. Nonetheless, the vehicle range we take for granted cannot be achieved sustainably, for a significant population, without hydrogen.
Hugo
