Neo1973: long-distance contender to Apple and Google?
Inside the mobile matrix, part 1
If you're a developer working with mobile devices few decisions will be more critical than the platform you choose to focus on. With the correct choice, your skills and creativity may pay off both financially and personally when thousands of users enjoy something that you brought into the world.
This being the month of the Consumer Electronics Show (CES) in Las Vegas, Nevada and Macworld in San Francisco, California, I’ve looked at three devices/development platforms that have tweaked developers’ interest: OpenMoko's Neo1973, Google’s Android and Apple’s iPhone. I will look at the strengths and weaknesses of each, to help developers decide which platform to adopt.
My first overview focuses on the Neo1973, the oldest of this trio enjoying renewed attention thanks to the hubbub surrounding Android and the iPhone.
First, some context. In order to make comparisons among these fairly diverse platforms, two of which are mainly device specific with the third being a more generalized operating system said to work across multiple devices, we needed some common criteria.
This process was complicated by the fact that virtually all of the information about the iPhone SDK and Android are based upon presentations, demos, documents and hearsay. With no hands on development time with the forthcoming iPhone SDK and with the first Android-powered phone release not expected until the second half of 2008 you might wonder how I can possibly do a review, let alone make recommendations.
The key is the questions asked. Instead of focusing on the nuances of each environment, I thought it more valuable to look at the factors that determine what a programmer can do with each device and how likely it is that the finished code will become sufficiently popular to have made your development efforts worthwhile.
Here, then, I shall examine the following:
- How big is the installed user base for the device or platform
- Does the platform limit you to a single device or can the code be easily ported to other devices
- What are the license terms and do they offer sufficient flexibility to be creative while allowing you to retain a reasonable interest in the work you've done
- How adventurous is the community that uses the platform or device
- How confident are we that this device or platform will be around for at least a few years
Is Neo the one?
The Neo1973 is named after the year when the first mobile call was made, and is designed to run OpenMoko.
Some people have compared the Neo1973/OpenMoko project with Nokia's Maemo/internet tablet initiative. Big differences exist, however. Starting with the fact that the Nokia device is a tablet only (it does not have GSM or CDMA capability), and extending right to the actual creation of the devices themselves. According to Nokia's vice president of convergence products multimedia Ari Virtanen "the N810 as well as its predecessors, the N800 and the 770 Internet Tablet are the only devices ever developed to be internet native from the ground up."
android isn't open, and it's not vapourware
open, when it comes to smartphones can mean two things; the device itself is open and all source code to make it work can be downloaded and built, or the SDK to develop application is available for building from source.
android runtime is closed and not all the SDK is open, so it's really just a free SDK. Symbian likewise! OpenMoko and the NEO are truly open, you can get pretty much every line of source excepting where regulators control it.
people have assumed that android is largely vapourware. well, it's not, you can download the virtual machine (arm processor) and run it on a PC (linux or windows) using the QEMU virtual machine container. Even better, you can extract the files and copy them to an Arm processor handheld such as a Zaurus and run it - even now, people are working on improving the hosting of android on top of either the Debian or Angstrom distros.
The Neo doesnt have to convince the FCC - it already has.
I dont know if you're from the US or not but anyway if a phone complies with the GSM standard and to local regulations (FCC in USA for ex.) - then one can sell and use it.
The Neo 1973 is approved by the FCC so you can use it with any telco using GSM and in fact with QTopia running on it you can use it as a phone, make and receive calls etc. (though being a dev-preview there are still some bugs to iron out). Walled garden or not.
I predict apps for the iPhone will come over iTunes ;-) and Apple will take a nice cut from the price. The only true open/free platform is the OpenMoko platform and it will remain open/free.
Accessing the phone - no surprise that it's proprietary
With all the talk around new phone platforms, this is the first time that I've seen it confirmed that there wouldn't be direct access to the phone module in any of the forthcoming products. I assumed that this was the case when Android was announced. Whatever the response of the FOSS mavens, this is a price worth paying for entry into the big wide world of mobile telecoms as it will make open systems possible up to the point where the call is being made.
However, I am increasingly convinced that the Neo1973 will never make consumer production. It's a proof of concept rather than a fully fledged mobile telephony platform, and will have to convince the FCC and other communications authorities before it even gets near the phone companies themselves, who also have their vested interests to consider, many of which don't include open source software. So what about Android? Google's might will push it through, as will concessions to the design to appease the communications gods. Most Android applications will be more like a widget running in a sandbox. It might not be as restricted as the iPhone, so you will be able to run ssh in a terminal over your GPRS connection, but restricted it will be. Mobile telephony has worked hard to build its walled gardens and won't give them up that easily