For the electric car to be a mainstream success it needs to follow the closed Apple model, not the open Google model.

Tesla Roadsters lined up outside of the Model S Beta Customer event

Tesla Roadsters

Updated: I’ve recently seen a rise in the discussion about the open source movement around electric cars, and cars in general. This article in GreenBiz (originally on Txchnologist) looks at an open source project for plug-in vehicles called Tumanako and the open source car design and manufacturing project Local Motors has been around for awhile. While these projects are all well and good (I fully support the open source movement in general), for the electric car to be a mainstream success it needs to follow the closed Apple model, not the open Google model.

If you’re not techie insider, I’ll briefly rehash the Google/Apple scenario. Google has long had a philosophy of making systems and software open and transparent, which in theory leads to the ability to tap into the minds of more and more developers — the idea is that the collective wisdom and passionate creators will deliver the best stuff out of the crowd. Apple, on the other hand, supports a closed system as a way toward better design and simplicity — Apple gadgets just work seamlessly, like magic, but they’re tightly controlled and mostly only work within their own ecosystems.

The mainstream electric car needs to support the closed, magic — it just works! — model. Why? First off, there’s little room for beta software mistakes in the automotive transportation world. An electric car that shuts down or freezes up because of half-baked software is the equivalent of a lawsuit or a PR nightmare.

Fisker Karmas

Electric car startup Fisker Automotive (which is not open source at all) has struggled in recent months partly because of software glitches, leading to some of its early adopter customers venting online, and a variety of negative press. The equivalent of Google launching a beta project onto the web, and the ecosystem of developers playing with and perfecting the tool, doesn’t really work for a mainstream car.

Yes, in a niche DIY community, open sourcing ideas about cars and electric cars is great, but I’m talking about the future of the mass produced electric car. The Nissan all-electric LEAF and GM’s extended range electric Volt are about the most mainstream, standard cars we have out there. I’ll also clarify that I’m talking about the design and creation of the car, not the ecosystem around the car’s data, which I think will eventually be an interesting and growing platform.

Secondly, for the electric car to go mainstream, it can’t continue to be pushed by the DIY, plug-in community, which is the equivalent of the open source tinkerer in the car world. Through our Green:Overdrive Show we’ve championed the DIY electric car folks, from putting a plug on a Prius, to stripping down funky cars and filling them up with batteries. And about four years ago former Intel Chairman Andy Grove compared the grassroots plug-in hybrid vehicle movement to the 1970s-era Northern California Homebrew Computer Clubs that paved the way for the personal computer.

But I hope the electric vehicle movement has finally started to move beyond that homebrew stage. Nissan has sold 27,000 LEAFs worldwide, with 11,000 of those in the U.S. GM hopes that it will soon start selling 3,000 Volts per month, with a record month of sales of 2,289 Volts in March.

Tesla is the electric car company that most resembles Apple’s model — make an electric car and a high-end brand with a cult-like following (yes, the following is still small) and try to protect the brand as it moves into the mainstream. Tesla held a successful IPO in the Summer of 2010, and its stock traded as high as $40 per share last month.

  1. Katie,
    You’re mistaking strategy and execution. Apple is not successful because they are “closed source”. They are successful because they successfully implemented a strategy that uses closed source. There are certainly a myriad of examples of closed source software that doesn’t “just work”; indeed, Apple has displaced a number of such vendors.

    Similarly, while Android certainly has some complexities, it is, nevertheless, remarkably more usable than most of its closed source competition / predecessors.

    While I agree that cars should “just work”, this isn’t a result of open source versus closed source. Apple’s products work they way they do because they are built by Apple, not because of the underlying licensing mechanisms.

    1. Yes, and to be successful, an electric car needs to successfully utilize a closed system model. You can’t just slap together a bunch of commodity parts and have a successful automobile. The best cars are those which spring from a single vision. Committee cars tend to be bland and don’t do well in the market. In any even, there is an intense amount of integration in developing and manufacturing an automobile. The more tightly managed this integration the better the chances for success. This is one thing that Apple also understands that is lost on many electronic gadget companies that fall into the commodity trap and can’t sustain themselves.

      1. Like any type of new product category or market, initially it has to be “closed” to succeed. As the market becomes more mature, and there’s a lot of know-how and customization, the market starts to “open”. Clayton Christensen calls this “integration” and “disintegration” in his Innovator’s Solution book.

        And it happens in pretty much any market. The original cars were not “open” either, but later did become open, because the advantage of integration becomes too small in the end, and customization is preferred.

      2. Local Motors’ Rally Fighter is a single purpose/vision that combines existing parts from other vendors. So far from reviews of the thing in action, they seem to have succeeded. The discussion of Apple’s markets vs auto markets is hard to compare, apples & oranges and all that. What you’re describing though sounds very much like the Apple vs IBM/PC market story which Apple nearly went belly up as a result.

      3. And I’m pretty sure all Redhat does is slap a bunch of free software together and give it to their customers without any form of QA or integration testing. Their number of developers is essentially zero, and all their releases suck.

    2. Exactly. Whether something works or not depends on the design and its execution, not just being open or closed. Historically, a lot of hardware vendors that go the closed source route put out a 1.0 and if you’re lucky a 1.01/1.02 software update before moving on and effectively abandoning your product. It’s even worse when the vendor goes out of business and there’s no hope for support or fixes of any kind. Open source software breathes life into hardware, keeps it running effectively, adding improvements and new features, while also keeping it from becoming additional obsoleted landfill.

  2. what about open sores?

  3. I have to disagree with Katie. Open source certainly leads to larger numbers of issues, but that is usually a function of larger use. The success of a product has everything to do with its maturity and support model and next to nothing to do with it being either open or closed source…

  4. The Apple model:
    – is more complicated than you make it out; it’s stacked on top of Darwin, which is open source
    – doesn’t always just work. iCal hasn’t been able to sync all-day events in New Zealand for years

  5. “There’s little room for beta software mistakes in the automotive transportation world.”

    Actually there’s *no* room for beta software mistakes in the automotive transportation world. The first car that crashed into another car and killed three people because of glitchy software would be the end of that nonsense.

    1. I hate this qoute, because for one it’s obvious. Two it’s implying Free as in Freedom releases are “beta software”. This leads me to believe the author has a slim idea how Faif Software works, and shouldn’t be writing an article to try to discredit it …

  6. Open source IS already happening, how do you think Tesla developed a DSP based digital motor controller based on the old Analogue AC Propulsion controller and got it up and running within a month? DSP chip vendors give out various examples of open source code to suit their chips and seeing as many vendors make DSPs specifically for the motor controller market, plenty of fully operational open source sample code is freely available.

    EV hobbiest have about as much chance of making a commercial powertrain from scratch as they do of making and marketing a DIY GSM mobile phone hardware themselves, so that part of the argument is WAY off track!

  7. Open source does not mean beta quality software. It simply means that people can see the software source and are free to improve on it. Open source is about giving many people the ability to analyze the code, discover flaws and report them. This increases software quality, and cannot decrease it.

    This seems to be confused with the idea of hobbyists installing their own software onto their machinese. People installing third party software, open source or not, onto their cars do so at their own risk and cannot hold the manufacturer liable.

  8. This article is very misleading. The truth is the most insecure & bug ridden software available to the general public is closed source. Instead of making it were everyone can see it and if there is a noted problem report it so it can be fix let’ take another approach. Let’s lock the code where no one can see it. So programers don’t have to be accountable and can do ‘sweep the dirt under the rug’ type programing. Just because it is open source does not mean it does not have a company or or payed employees behind the project. It just means it’s all out in the open, instead of hidden.

  9. Oh my goodness. And this author is a technology writer?

  10. I don’t think the author understands what open source vs closed source development model is!

    Apple is successful because it has pretty good quality control, regardless of its closedness (secretly they run atop a LOT of open source code, developed largely by 3rd parties).

    Google promotes open source development initiatives, and their quality control is good too, for Android, for example. Truth be told, Google does not handle Android like a usual open source project. They work on it behind closed doors and once they deem it ready they open it up.

    Open source development is about sharing and contributing ideas and together improve the code/hardware. This has been traditionally how hobbyist handle their projects. There is no concept of ‘patent’ here, just that of sharing information. Companies prefer the closed source model because they want to squeeze every penny out of their work via patents and copyrights and prevent competitors from ‘stealing their ideas’.

    Open source initiatives may not be as productive and may have issues at times because they lack manpower and resources. If they had the financial might of GM behind them, I think things would move along much better. Many companies are embracing and succeeding with open source development (e.g. Red Hat), but the movement remains widely unknown because the big name companies want to keep the status quo… it the end it all boils down to morals, which are lost in corporations, where profit (for the shareholder) is king.


Comments have been disabled for this post