Lies, Damned Lies, and Browser Statistics
DSL Reports is running a story about Ars Technica is seeing IE popularity plunge. Ars Technica is kind of a geek site, so I would expect that it wouldn’t give you a true picture of the browser preferences of the average user. I’d be interested to see what a site like Yahoo, Google or eBay gets in terms of browser users.
Even so, here’s why I take most people’s quotes about browser statistics with a grain of salt:
- Most non-Internet Explorer browsers can be made to give off a different “User Agent” string. The reason browsers need to do this is because some web sites use the User Agent string to allow access to a site (e.g. the site only works on Internet Explorer). The sites may, in fact, work perfectly fine on a different browser, but the site is stubborn and won’t let you in otherwise. The end user makes the browser “Lie” about who it is (e.g. pretend it is Internet Explorer on Windows when it is really Mozilla on Linux) so you can get into the site.</p>
- Sampling methods are spotty at best. It’s not as if large sites like Yahoo, Google, or eBay are going to share their web logs with some third party to see what the real trends are in terms of browser usage. Also, every site has a unique audience, so one site might see one trend (e.g. IE over Mozilla) whereas another site with a focus on non-Windows platforms would definately show more end users using Mozilla than IE.
- Almost everyone who uses an alternate browser (i.e. Not Internet Explorer on Windows) also has to use IE on some site or another. There’s a part of the eBay site where you must use Internet Explorer due to some bug somewhere.
- During the time that IE was gaining marketshare on Netscape back in the mid 1990s, I actually worked at Netscape. I remember us essentially challenging Microsoft’s claim that they had taken market share from Netscape. Our rebuttal was that we still had 70% market share. Not sure how the two companies got their numbers, but when two companies disagree so wildly on a statistic, someone’s gotta be making shit up somewhere.</ul> That being said, I do believe that the tide is turning away from Microsoft and Internet Explorer. How much it is doing so is debatable. I’m happy to report I have done my small part. I’ve had my wife running Mozilla and/or Firefox for quite some time now. Having recently fumigated my inlaws laptop of spyware and the like, I used it as an oppotunity to install Firefox on their computer and show them how to use it instead of Internet Explorer. When I told them “you’ll get less spyware if you use this browser,” it only took a couple of seconds to make that decision.
- Almost everyone who uses an alternate browser (i.e. Not Internet Explorer on Windows) also has to use IE on some site or another. There’s a part of the eBay site where you must use Internet Explorer due to some bug somewhere.
- Sampling methods are spotty at best. It’s not as if large sites like Yahoo, Google, or eBay are going to share their web logs with some third party to see what the real trends are in terms of browser usage. Also, every site has a unique audience, so one site might see one trend (e.g. IE over Mozilla) whereas another site with a focus on non-Windows platforms would definately show more end users using Mozilla than IE.
