Does that matter to us? When I'm deciding what to support as a developer I care only about the size of the platform.
back
5 comments
If all you care about is number of potential users you should target either S60 or JavaME.
Or HTML/JS.
"Does that matter to us?"
Yes, I imagine fragmentation would matter when it comes to developing for a platform.
Indeed. Fragmentation is probably the biggest reason we'll never see Android Market-sales on a per-device basis top what each iPhone user spends in the App Store- there's a wife range of CPU power, RAM, graphics capability, peripherals, multitouch or not, etc on Android devices. Like being back in the world of Windows CE...
The same is happening to the iPhone ecosystem; at a slower pace, but it's happening. Before it was the iPhone 2G; now it's iPhone 2G, 3G, 3GS, iPad, soon the new iPhone 4G... soon there will be way too many differences in resolution, screen size, CPU and OS too.
Sales per-device is irrelevant, IMO. If I sell 100 for 10 devices, it's better than selling 90 for 3 devices. It's not like I have to add one IF() condition for each new model in my code; merely that I have to plan for slower CPUs and different screen sizes and resolutions, and when I do so, it's for a whole group of similar devices.
It's a bit different though -- if you look at stats on Flurry for some of the popular apps, you'll see that there is less than 3% iPhone OS 2.0 users out there. That's simply because iPhone users get a free OS upgrade when they hook up to iTunes. The majority of users who are still stuck in the 2.0 land are iPod Touch users since they actually have to pay for the 3.0 upgrade. That's why the iPhone market has significantly less fragmentation.
However, I have a HTC MyTouch android phone that is still stuck in 1.6 land. I upgraded from 1.5 recently, but it was not the simplest process to do so. HTC doesn't really make it easy for phone owners to upgrade their phone OS. It's possible for me to use the CyanogenMod to upgrade my OS to 2.0, but not without lots of effort at the command line.
The current slew of 1.5/1.6 android devices out there right now will probably never be upgraded to 2.x mostly because the phone manufacturers rarely make it an easy process.
People don't seem to notice that on the iPhone platform fragmentation is limited because people keep buying new devices.
Usage stats from my app in April (somewhat unscientific but illustrates my point decently): iPhone 3G: 49 users iPhone 3GS: 48 users iPod Touch 2nd Gen 4 users iPod Touch 1st Gen 3 users iPhone Edge 2 users iPod touch 1 user iPad 1 user (might have been me...)
Well, the 2G is being phased out with the next OS release, meaning that even though people will still use their old one development will still be focused on a few core devices. The first major fragmentation will be with the new hi-res screen on the next iPhone, the 3G and 3GS aren't really that different to develop for.
I think you would care about making money more than the size of the platform. There was a time when MySpace had 2x more users than Facebook, but app developers on MySpace didn't have the same level of virility that Facebook apps had and thus it was far more lucrative to make FB apps.
I agree. Which is why those numbers aren't that interesting when comparing Android to iPhone OS, since iPod touches and iPads are a part of the platform.
That being said, it clearly says that the Android platform is growing.
It's not about the size of the market as much as the quality.
Would you rather have 100,000 potential users and make 1,000 sales or 200,000 users and make 800. Or maybe make the same amount of sale but being able to charge more because you know that who spend $500 for a phone is more incline to spend $10 on an app than somebody who spent $100?
Ok, but that is a pile of assumptions.
If the iPhone user buys 3 apps for every 1 the Android user does, but in a couple years there are 10 times as many Android users, where do you sell your app?
The attach rate and the market size make all the difference, and there is no data to support the conclusion you are trying to make.
A gaming PC is a hell of a lot more expensive than a console, yet their attach rate is way way lower than a console.