Older: 
Android/G-Phone Software Development Kit - First Look
Location: Ottawa, Canada
On Monday of this week, Google and a few other folks (actually a pretty wide range of semiconductor companies, handset vendors, cellular operators and others) announced the long-awaited G-Phone (Google Phone). What’s interesting, however, is that it really wasn’t an announcement about a phone at all, but rather about: an alliance called the Open Handset Alliance; some software that might constitute an operating system based on an open interface; and a software developers’ kit (SDK) to access the functions of the phone and, possibly, the mobile network for the purpose of enabling flexible applications development.
This announcement rivals the hype of previous “earth shaking” technology such as the Segway human transporter. Because we didn’t know what to expect and because the parties involved were pretty smart people, we expected the world to change with the publishing of a press release or an interesting name. It didn’t.
Although the announcement this week has the potential to change the landscape of mobile networking, the reality of how impacting it will be is still up for debate.
A few initial observations…
First, the idea of an operating system that allows applications to be created on cell phones isn’t particularly new. Qualcomm has something called BREW, Symbian exists, as does Windows Mobile. Granted, they may have different interfaces and cost models but, in essence, they all provide the same basic functions. What may be different with Android (the Google-driven SDK) is how it interfaces with the web-based services that Google offers (Maps, Mail, Search…), but since the SDK won’t be available until November 12th, I will wait to make that judgment.
Second, just because a phone has an interface and operating system on it does not mean that the network it is connected to freely presents all the capabilities that that network can provide. In fact, most network services are controlled so as to create revenue streams (billing for minutes, for example…) or to protect the privacy of consumers (location details, for example). Changing the API or even the operating system may not change any of this, although it may allow for easier access to the services that can and should be presented. Until we see a few devices with this software implemented we can’t know just how much will be different.
Third, while the idea that a free and open source operating system should change the total economic equation, the actual impact is hard to calculate. For example, if you add up the cost of a smart phone, battery, core chipsets and processor, baseband chips and memory, along with screen, plastics and other components, they may very well cost far more than the OS. A free OS probably will not change the economics of the device significantly. See the iSupply teardown of the iPhone for an example of the cost structure of a smart phone.
Fourth, the assumption that there is a lack of applications in mobile devices might be overstated. While I am the first to agree that a more flexible model is needed, there have been a fairly large number of applications already created for mobile devices. Just look at the number of third-party applications available on a Blackberry, Windows CE phone, or a host of other devices. What I will say, though, is that the major differences between these OS’s creates a problem in the sense that an application built for one usually must be recreated for the others using specific SDK’s and API’s. This complexity will probably not be made less significant by adding another SDK, unless Google can also drive compression of the interfaces used and catalyze greater standardized interfaces. I would certainly not count them out in terms of being able to do just that, but time will tell.
Finally, in my opinion, the issue with mobile devices is not the absence of presence of applications but rather the fact that the cellular experience is not yet equal to the experience we get over other networks, such as home networks, enterprise networks, and even WiFi. In that regard, the single most significant investment we can make in revolutionizing the mobile experience is to invest in next-generation networks - such as 4G technologies like Mobile WiMAX 802.16e and Long Term Evolution (LTE) - or in intermediary 3.5G systems such as CDMA EV-DO Rev A. Of course, I am biased since Nortel creates these networks, but to me it is clear that the thing that catalyzed the wireline Internet was not just a better SDK, but rather the abundance of low-latency, high-capacity, low-cost bandwidth. I expect that as we scale the capacity of mobile networks, the capacity, not the API’s, will be the major driver of innovation and new experiences.
Don’t get me wrong. I like the Android effort and the idea of anything that creates more devices to connect to the networks we create. I also love operating systems that make new applications possible so they can consume more sessions, minutes, bandwidth, conferences, and network intelligence. And, I can’t wait to dig into the SDK once it’s finalized and formally out to see what innovation it spawns. What I am sure about is that the focus and interest this announcement has created is a good thing for mobile networking, which is a core competence of Nortel.
