Attracting Mobile Application Developers

Google’s “Open Source Engineering Manager” Chris DiBona was recently defending his employer’s rather dismal open-source efforts on its supposedly open-source mobile operating system Android. He admitted that they should do better, but for the moment Google is totally focused on gaining market share to attract application developers, and isn’t too concerned with building an open-source community. He re-enforced several times that Google’s strategy for attracting mobile application developers was “market share and device volume”.

Although I appreciate the way Google keeps their strategies simple, this one struck me as misleadingly simple, so I decided to write-up what I believe the key success factors for attracting mobile developers.

1. Maximise revenue opportunity
   1. Device sales
   2. Low fragmentation
   3. Effective sales channel
   4. Consumer demand
2. Maximise competitive attractiveness
   1. Market share
   2. Unique features
3. Reduce product development and maintenance costs
   1. Low fragmentation
   2. Development tools, documentation, support
   3. Skills supply (i.e. lots of people who can develop for the platform)
4. Awareness and perception of all the above factors

.

Looking historically across smartphones:

• Symbian/Nokia focussed primarily on (3) and (2) (and did reasonably well at them). They were extremely poor at both (1) and (4).
• Apple completely nailed (1) and (4) and did enough on (2) and (3). The rest of the industry is now benefiting from Apple’s efforts on (1.4) and (4); we really needed a company with real consumer marketing competence.
• JavaME did OK at (3), but poorly on the rest in general. Some network operators did quite well for the time on (1) (e.g. NTT DoCoMo and Verizon) and had some successes.
• Microsoft did their usual excellent effort on (3); but focused on the enterprise, for some reason hoping that their PC strategy would work 15 years later in another market, so they failed on (1), (2), and (4).

.

Looking now at Android — they’re doing mediocre on (1.3), (1.4), (3.2), and (3.3); they’re probably exceeding on (4); and they’re not doing so well on (1.1), (1.2), (2), or (3.1). Given this, focussing on device volumes and market share is probably the best objective, but they need to be worrying alot more about fragmentation, and at some point they’ll need to do a lot better on sales channel.