The overall context of the discussion seems to have been focused on phones. At least, that is the impression i got from reading it.
Well, then maybe what should get invalidated is your impression.
Actually, its not. For example, one could easily imagine that browsing is far more common and that people spend more time browsing while doing it on a tablet, than on a phone. Given the numbers of iOS v. Android tablets, one could then quite easily conclude - given the premise - that such a result is to be expected.
Yep, one could easily imagine lots of things.
But I prefer referring to the real numbers.
In
http://www.netmarketshare.com/brows...=1&qpcustomb=1&qptimeframe=M&qpsp=132&qpnp=11 you can see that the number of iOS browsers in the wild didn't change significantly on January 2010 (introduction of the iPad). Since then, Android has surged in selling numbers to the point of dethroning Apple not long ago.
That should be (but isn't!) reflected in the graph for this year:
http://www.netmarketshare.com/brows...=1&qpcustomb=1&qptimeframe=M&qpsp=143&qpnp=11
So: your theory was that iOS devices includes tablets, so that explains the (MUCH!) greater proportion of iOS browsers being detected. But as shown, statistics from before the introduction of the iPad and the evolution of said statistics suggest otherwise: looks like iOS users, tablet or not tablet, end up using their devices as web devices
astonishingly more frequently than Android users.
(one thing I don't understand is why that seems so... painful? to you; to me just is something interesting, food for thought)
Which leads us back to my original point. The data itself is quite useless. At least in the way it has been used in the discussion thus far.
You see, some interesting things can be seen if you are careful. Or willing.
I can think of several more. Watching videos. Playing games. Chatting. Listening to music. Browsing catalogues. Reading newspapers (Go mag+!), and the list goes on.
Yep, or buying them in bulk and using them as ballast. (Hey, it's
possible!
)
While some of these in one way or another may rely on data transfered over TCP/IP: a) data is not necessarily proportional to use, and b) data is not necessarily captured as part of the analysis - for example, if i wrote an app that connected directly to my content server, which was then used to pull (read: stream) massive data sets to the client device, how would the survist know? (answer: most likely, they wouldnt).
You're right; as an example, data transmitted over a BTP (Ballast Transfer Protocol) wouldn't appear on any statistic.
Which, as mentioned above, could for example be explained by iOS dominant position in the tablet space. Not saying that is the answer, but it very well might be. Point is, it (the data alone) says very little (read: nothing) about a) Androids capabilities, b) how said capabilities are being used by the end user, and c) why this is so.
It doesn't say anything about (a) nor about (c).
But about (b), it DOES say, clear and loudly, that iOS devices are... (let's repeat once more...) MUCH more frequently used to browse the web than Android devices.
Meanwhile, maybe Android users are somehow, for some reason, careful not to do so.
Sure, but ill stick with the shirt. Shirts are better when going t-shirt ninja!
A bib will it be then.
Didnt know there were such a thing as "a real smartphone", or a defined way of using one. To each his own. Thats the beauty of (somewhat) general purpose computing, isnt it?
Well, I tend to think one important difference between a smartphone and, say, a piece of ballast, is that one is used to browse the web and the other isn't. I'll let you decide which is which.