All Things Web: Debunking Browser Stats
Lurking behind the invalid, inflated stats that invariably accompany claims that “99.999% of users use Browser X or Browser Y” is almost always an implicit statement:
“I want a high-sounding excuse for exclusionary page design.”
But browser stats make a poor basis for Web design. Remembering that not all “Netscapes” are created equal, the most widely-used release of Netscape for Windows accounts for roughly 20%-30% of Websurfers. (Let’s split the difference and call it 25%.) Not 80% — but 25%. The world’s second most popular browser accounts for roughly 10-12% of your potential audience. Thus, a visually-driven design that looks great in both browsers covers about a third of your readership.
No matter how you slice it, there’s just no basis for using reports of browser popularity as an excuse for exclusionary design.
I am dealing with this daily right now as I work with page templates that look great in MSIE, or more correctly, Microsoft Visual Interdev, but flat out refuse to render in anything else. Sadly, they parse as valid HTML, so it looks like a bug in the Gecko renderer, but that doesn’t excuse not testing it.