anyqs
How did they calculate the percentage?
As far as I know there is no accurate way to determine exactly the percentage the market share of browser. So, I want to know their standards on their way of counting.
Besides that, what does the empty space represent?
Moreover, do they include all kinds of devices (laptop, phones, TVs, desktop, ...etc) or just laptops? And if so, how did they distinguish between them in their study?
:)
I'd like to imagine that mobile browsing was done mainly on iPhones as at 2009, the latest date on the graph. Did that contribute to Safari's market share? If it did, were other mobile browsers counted as well? Also, how exactly is the data gathered? It mentions w3schools, the web consortium, but doesn't mention how they (w3schools) get their data or from whom, and what kinds of devices are polled.
anyqs
I would like to see the more information on how they collected this data and what it applies to. As James said, mobile browsers are distinguishable from others (from something called the User Agent in the browser), but I could go into my Opera browser on my Android and change the User Agent to 'desktop' and trick the website into thinking that a desktop viewed it. I would imagine that if w3 were to pull from their website, it would be heavily biased towards non-mobile browsers, as people would prefer to learn on a more comfortable screen.
This was rather interesting because, although there is sampling bias, it doesn't really matter. The sample space is android devices that visit the android market. Here, the sampling bias means only devices that would use the android market are polled--which overlaps very well with the developer's target audience of people who might download their app.
This is a sister website of the very popular www.speedtest.net, which does many millions of internet download/upload tests every month. Aggregating that data across the world provides some pretty robust statistics on regional and national connectivity speed. You probably never knew South Dakota had the second highest average download rate in the country.
http://www.netindex.com/download/2,1/United-States/
I'm guessing that's related to their cheap power and ample space in encouraging data centers to move in.