Page 2 of 2
Almost an Android
What is surprising is how little modified the Fire version of Android actually is. Amazon gets off the hook in not having to create a development environment. To start work on a Fire app you install the same SDK as for an Android app - just remember to select the version used by the Fire devices you are targeting.
The Amazon SDK provides some extras such as an emulator for the Fire - there isn't one for the 8.9" device at the moment - and some code to connect to the specifically Amazon services.
It is mostly the Amazon services that distinguish Android from Amazondroid - these are:
- Amazon In-App Purchasing
Google's in-app purchasing doesn't work on the Kindle Fire. If you want to allow users to buy digital content and subscriptions you need to use Amazon's own in-app purchasing API.
This API is optional but it can make your game more appealing. Users can sign up and have their scores added to public leader boards and earn achievements and so on.
- Amazon Maps
Amazon has signed on to use Nokia Maps instead of Google maps. The API is supposed to be easy to migrate to from Google maps but as the API is by invitation only it is difficult to confirm this. One problem is that the first generation Fire doesn't support the Maps API.
These are the three big services that you are going to have to consider when you port to Amazondroid, but there are some other more minor issues:
There is a built-in email client that responds to mailto and email-intents.
- SD Card
The Fire tablets have an internal SD card that your app can write to - this is not removable or upgradeable. It should work with no changes to your app.
A strange feature of the entire Fire family is that they all support Adobe AIR. Version 2.7.1 is pre-installed on Kindle Fire (1st Generation); Adobe AIR version 3.1 is pre-installed on Kindle Fire, Kindle Fire HD 7", and Kindle Fire HD 8.9". What this means is that as well as Java you can use ActionScript.
This is a great opportunity for any AIR apps that are seeking new devices to run on but given the way Adobe is abandoning Flash and related technologies it probably isn't a good bet for new projects.
At the end of the day porting an Android app to Amazondroid doesn't look too hard. However, the fact that porting is needed at all is a great shame and, yes, it is fragmentation.
In the real world, however, the fragmentation shouldn't stop apps from being published across the entire Google/ Amazon Android family - things could be worse.
The good news is that we can increasingly focus on Android 4 apps.
It would be better if Android was a single platform and preferably with over-the-air auto-upgrade to the latest version, but that's utopia. For now, we have to face the fact that both flavors of Android are worth targeting, as Amazondroid shouldn't require much extra work.
The final problem, however, is the simple fact that we have no idea what the actual size of the Fire market is. Amazon has been careful not to let slip its sales figure. Recently it claimed a 22% share of the tablet market, but this seems unlikely to be a meaningful statistic. Independent estimates from IHS iSuppli put the figure at more like 14%, with sales falling in recent months. IDC put the global market share at less than 5%. Amazon's claim to have sold 1 million Kindle devices each week might well be true, but it doesn't indicate how many of these were the bottom of the range ink devices. Put simply we still don't know the size of the potential market for Fire apps.
The bottom line is that developing apps that target the Fire range and sell from the Amazon App Store is a good gamble.