The main thing to consider is that browser usage statistics can vary wildly depending on the website. In the case of the statistics above, they are taken from a large sample of websites, but the catch is that you need to sign up with W3Counter to get your website tracked. Subsequently, these websites are unlikely to reflect the same sort of user base as ours do. By way of an example, less than 0.1% of visitors to Mizuno Golf use Firefox 1.x, and this figure is very similar for most of our clients we monitor statistics for.

In short, the numbers above do not reflect the web as a whole, just a very specific sample. This sample is also likely to be a bit more techie than most, hence the high-ish number of Firefox users (which is actually double that which we would expect on most of our websites).

It should be noted though that while we don’t actively support certain browsers, this doesn’t mean our websites will not work in them – if built to web standards I would expect Firefox 1.5 and Safari 2 to render pages accurately. We just don’t physically test in everything, but we do ensure that our websites will work in our supported browser list, which accounts for approximately 98% of our end users.

Hope this sheds some light on things! :)