The Londroid September meetup was a great success. Here is a quick review for people who could not make it.
Kabbee – A marketplace for minicab journeys – Phil Makinson
What is Kabbee?
Kabbee is a price comparison and booking service for quality minicabs in Greater London.
Very nice use of a smartphone. This is a disruptive application, it provides discovery, a market, transactions and it is scalable.
Nothing too clever about the front end and back-end technologies although they may perhaps have some competitive advantage from their proprietary implementation of 3D secure on mobile for credit card payments.
I wish I thought of that.
Roboguice – Dependency Injection on Android
Michael Burton, the creator and lead of Roboguice, will be gave us an overview of the library and how dependency injection can save you from writing extra code.
A really good talk. If this is the sort of thing you like, you will like it. Last year the 300kB download and performance penalty due to reflection might have mattered, on next year’s devices you won’t even notice.
So you don’t want to go native – Kieran Gutteridge
Kieran gave a very interesting talk about Smartphone development. He have been doing it for over 10 years and is personally familiar with most of the OS and tools.
It is always invideous to summarise someone’s 30 min talk into a few lines but my takeaways were:
- porting well written iPhone apps to/from Android is easy. The UIs will be different on both platforms but the back end code remains much the same.
- There is money to made from J2ME applications particularly in Asian markets. This is because getting your app into carrier’s stores is so hard there is little competition.
- Write once – run anywhere apps like Titanium and Phonegap “don’t feel right”. They are fine is the user is meant to understand them as encapsulated websites but so so good if the user is meant to understand them as native apps.
- There is real fragmentation of browser capabilities on Android phones today. This makes it difficult to deploy powerful HTML5 applications.
From Hobby to App ‘Success’ – Chris Banes
Chris Banes is an Android poster child. A corporate developer by day he coded Android as a hobby in the evening. His app was hugely popular, made lots of money and was bought by a big company who then employed him to work on it. It now has 1.5 million downloads. I bet his girlfriend is hot too…
Before younger and more impressionable readers think “I can do that too” let me explain further. Chris developed a Facebook client for Android. His was first and best and he started hitting the Facebook API a couple of days after it was first published. He then had the courage to quit his full time job and go for it.
Chris did not succeed because he was a better coder than others (although he might be) he succeeded because he was in the right place at the right time and that requires far more skill.
Chris explained that ~50% of his revenue comes from advertising. This is nice because it is a recurrent revenue but it only works because the user spends hours staring at his app. It will not work so well for utility apps such as my own Keyapt where the user only glances at the screen.
Overall, Chris’ experience does not dissuade me from my view that the app market is broken. There is oversupply and users pay too little for developers to be able to produce high quality software and a reasonable living.
I would love to be proved wrong so if there is anyone out there who pays their mortgage and supports their family on their earning from their own apps please tell me and I will give you a plug.