> Any thought?

I guess I'd be somewhat wary of comparing general trends to a more
defined population.  I'm guessing your campus population is not
typical of the national population, instead probably skewed towards a
younger population with higher disposable income (and also perhaps
more sensitive to peer pressure) and hence might not follow general
trends ;).

Also, how is your 70% traffic figured?  Do you have any way to
determine if perhaps a few outliers are creating a significant amount
of traffic.  (In other words, do you know if the mobile traffic
actually represents ownership, or might there be a smaller group of
i-phone users who happen to use the library services more? I'd guess
the smaller the population accessing via mobile, the more likely a
small population could skew the results)

Also, how are you measuring the Android users?  Is it possible you're
missing some who would be using non-default browsers or browsers
modified by a carrier?

I don't unfortunately have any stats, but I do seem to remember seeing
some numbers locally that would indicate iOS count of web usage is
still pretty high. Android phones are becoming very, very cheap but
data plans aren't.  Also, the form factor and the processing power of
some of the cheaper androids make web searching less than thrilling.
I could see someone using an Android that they get for free, but not
accessing the library for a variety of reasons.

It would be interesting if one could compare the usage of different
Android devices but the difficulty of data collection here might be
enormous. (I'm not sure off hand if there's an easy way to
distinguish, say, a Samsung Galaxy 2 from a Optimus)

Jon Gorman