The Android platform is very fragmented in terms of OS version deployed, OS extensions for different hardware, and hardware capabilities. It is important that The Graphics Codex software runs efficiently and reliably, and Android is a testing and support nightmare (as is Windows for desktop software). For iOS, I have three physical devices plus the simulator. I invest a lot of time ensuring that the experience is as good as possible on all of them, with iPad 2 on iOS 6 as the lead platform. Apple has a single OS version, relatively few form factors and processors, and a single point of contact with readers–the App store. On Android I’d need many more devices (I’ve been testing on a Lenovo tablet so far)

Unlike an e-book, The Graphics Codex actively performs layout, image resampling, and some interactive computation (e.g., for filter equation plots in the upcoming release). Any Android 4.x+ device can support these features the way that I’ve implemented them, and Tegra 3 devices have sufficient GPU power to perform them well. As of this week, about 25% of Android devices are running 4.x+ (http://developer.android.com/about/dashboards/index.html) I do not know how many of those are running Tegra 3. Unfortunately Android is currently the Linux of mobile for graphics developers; a lot of us like the platform and advocate for it, but it is hard to ship a product for it. And much like Linux, there is no technical barrier for me in releasing a product on it at any point, but it would be unfair to release a product that I wasn’t prepared to support adequately.

I’m very happy that I’ve been able to respond to every bug report and suggestion for the Graphics Codex, usually within 24 hours and a patch distributed as soon as Apple’s review process allows (about 2 weeks). I won’t release the Android version until I believe that it will run well on every compatible device and that it is feasible for me to support those devices at the same level I have been for iOS. The good news is that Google seems to be exercising tighter control over Android and that more Android devices are shipping with good processors than ever before. If the Google Play store becomes the distribution channel of choice and Android users more consistently update to the latest OS version, then this could become a viable platform for the next Graphics Codex port. Although nobody has asked, right now Windows 8 isn’t on my development roadmap at all. That platform has good control and standards, but is a minority in terms of market share.