Freeing your phone: Q&A with Sean Moss-Pultz, president of OpenMoko, Part I

		中文繁體 \| 簡體

HOME

DIGITIMES >> Digital CE magazine

Eric Mah, DIGITIMES, Taipei [Tuesday 4 December 2007]

Google's announcement of Android in early November brought a lot of press to the idea of an open handset platform. Such an idea, however, is not new and OpenMoko is one company that has been publicly pursuing such a goal for more than a year. Digitimes had the opportunity to talk with Sean Moss-Pultz, president of OpenMoko, about the latest developments at the company and what effect Android will have on the project.

This is part one of a two-part interview. Part two will be published on December 6.

Q: How did you become involved with the OpenMoko project?

A: About two years ago I was sitting down to define the product roadmap for mobile phones for a new group in First International Computer (FIC). We were going to make smartphones using Windows Mobile, but I kept thinking about how we were going to differentiate from among the other 20 plus companies in Taiwan making mobile devices. I stepped back and asked, "If you could start from scratch, if there was no legacy code, what would you do to make a difference in the GSM market?" Then it kind of hit me, it needs to be an open phone, it needs to be the one thing that the phone has never been, and that is totally open.

I like to describe the mobile phone as the most personal electronic device we have, since almost everyone carries one. But, it's also the most closed piece of electronics we have. OpenMoko is out to change that, to make the phone whatever you want it to be.

Q: Why GSM?

A: GSM has two things going for it. First, the market is incredibly saturated – it is by far the largest wireless market in the world. Second, it works off a SIM card. A SIM card is standardized, so regardless of where you are in the world, you can buy this phone and pop your card in and it'll work.

Q: How many people are working on the OpenMoko project currently?

A: OpenMoko was officially spun off as an independent company in September of this year. Internally, we have roughly 50 people at OpenMoko. On the community side of things, it's hard to say. We have about 10,000 people who subscribe to our community list on which we discuss everything about OpenMoko. In terms of actual developers working, I'd say we have about four or five hundred people. The project is still in its early stages. We have been shipping a developer version of our product for almost six months, so people are still learning what this is all about and how to use the tools.

We planned for our first customer all along to be the developer and from the developer we will move slowly to other people. Our focus is to get developers excited about this product, to morph communications and make the phone more of what it hasn't been before.

Q: Can you define "developer" a little more here?

A: Initially, a developer is someone who is an open source hacker, but that's not our ultimate target group. Our target is anyone who has any interest in programming, even someone who may not have done programming, but who wants to get into the phone and understand it and change it. Because we are still developing the platform our tools have not been user friendly enough, but next year you'll see a huge focus from us to make these tools easier to use as we try to lower the learning curve.

Q: Just how much is FIC involved in OpenMoko? What kind of support is the company providing in terms of financing and human resources?

A: When we became a privately held independent company, FIC became our primary private investor. FIC does the manufacturing and they have helped us out administratively in getting started.

Q: What are some of the major milestones OpenMoko has hit since the announcement of the project and have there been any problems along the way?

A: OpenMoko was announced in November 2006 and we opened up code for everything in February of this year. Most products have the device first and if it's good a community forms around out it. We did the complete opposite; we built the community first and then produced the device, so that was interesting from a development perspective.

Though we had some sourcing issues for some major hardware components, which delayed the shipping of hardware until early July, the start of shipments was another major milestone.

The GPS chipset that was originally in the phone needed to have a closed source binary library to run in user-space and the processor we had at the time was a bit slow and didn't have graphics acceleration, so we did a redesign. We put in a GPS chipset that was completely open source, a faster processor, Wi-Fi and other things. This is perhaps the biggest milestone for us up to this point, because we now have a device that is not only powerful, but also totally open. There are no more struggles for us to figure out the right licensing terms to ship binary code.

Q: What GPS chipset are you using now?

A: We're using chipsets from a Swiss company called u-blox. Before we were using Global Locate but then Broadcom bought them. We got the test results back for the current GPS unit and the performance is phenomenal. The performance is as good as the best car navigation GPS devices.

Q: At what stage is the OpenMoko project at? There are reports that the Neo 1973 (OpenMoko's first handset) will be available to consumers by the second half of 2008. According to the schedule on the project's website, it seems that things are still in the Developer's Preview stage. Any updates on this?

A: We are at Phase 1 of our development in which things are open to hobbyists.

Like I mentioned before, we did a hardware refresh of the device so it now has graphics acceleration, Wi-Fi, a faster CPU and accelerometers so you can do tilting things with the device. The design is still the same, however. The actual specs of the device include a 2.8-inch VGA (480×640) touch screen display; Samsung s3c2442 CPU running at 400MHz; Bluetooth 2.0; Wi-Fi 802.11b; TI GPRS 2.5G; AGPS; USB 1.1; 2.5mm stereo jack; 128MB SDRAM ; 512MB NAND flash; NOR flash (for recovery); S-Media 2/3D graphics accelerator; and two 3D motion sensors.

The most interesting thing in the phone now is the Wi-Fi because it opens up the whole concept of free-networks as well as freeing the phone. There were a couple of delays because of the Wi-Fi driver but everything is OK now.

We will push everything into mass production in mid-December and will be shipping in a few months after that. This will be more workable hardware and is consumer-ready hardware. The software stack will be more complete as well but that is still mainly for developers. But the Neo is now finally a device that's not going to change at all when we are going to ship it to consumers.

The base kit of the current model, which is just the handset, is selling for US$300, while the advanced kit, which comes with a debug board, is selling for US$450.

Q: Who has been buying your devices so far?

A: Basically, buyers in the United States and Europe account for 85-90% of our current sales. I hope we can get people in Asia more involved and excited about it but basically up to this point it has been a US and Europe thing. There are a couple of reasons for this. First off, most public open source development takes place in Europe and North America. Secondly, this is kind of a hobbyist device right now and I think this fits in better with the culture in those areas.

I'm really hoping to change this, to encourage use in Asia. We are trying to do a lot of things here like setting up an Open Lab to encourage people to come in and work with us, so I'm hoping to see an increase in sales from Asia in the next couple of months.

Q: You've opened a lab where the public can come in and use your facilities for free?

A: We're making a device that you personalize, that you make it the way you want it to be, and so we wanted to make it easy for people to come find us. On November 17 we opened up what we call an Open Lab here in Taipei so that the community can come and use our computers, talk with us, and look at what we're working on.

Q: What are your plans and goals for this Open Lab?

A: We'll be targeting universities and schools here in Taiwan first. Most of the open source development here is focused at the universities. If this model is successful, we may reproduce it elsewhere. It's very experimental. We have no agenda. We just want people to come in and try things. It's our way of trying to better understand what people want.

Q: What do you think of the Open Handset Alliance and the recently announced Android handset platform? What makes OpenMoko different from Android?

A: First of all, we are a product company that is developing a software platform to help make great new products. We see our strength in two places. One, we want to build a brand, a product that is exciting for a niche market. The second is our ability to integrate software into hardware. Both of those are very difficult to do, but if done correctly have an infinite potential to build value.

Android is a platform that is not supposed to run on any specific hardware. Because it's such a phenomenal amount of work to get a kernel to run optimally in an embedded device, we'll have to wait and see just how well Android will work. I can't make too many comments on it because I've probably only seen what you have already seen, which is the SDK. There is no phone that runs Android right now. We, however, have been shipping a phone for six months already.

As for the Open Handset Alliance, we at OpenMoko have decided to not be involved at all, initially, in any kind of standards bodies. We try to build devices in the spirit of the open source world, which is, you don't write standards, you write code and the code becomes the standard if people like it. We're just going to get things done. We're not going to sit in meetings and talk about how things should be. We let the community tell us what they want and we just do it.

Part two of this interview will be published on Thursday, December 6.

Sean Moss-Pultz, president of OpenMoko