Re: HTML5 vs content type sniffing

Ian Hickson wrote:
> To make a fair assessment of upgrade adoption rate, you have to discount 
> IE7 installs due to bundling with new computers, and you have to compare 

Why? I'm not sure why it would be relevant why somebody uses a specific 
browser. What counts is that he or she does. If I replace a Win2000 
machine with IE6 by a XP machine with IE7, and now use IE7 instead of 
IE6, how is that not "adoption"?

(My new desktop machine came with IE7 bundled, but that will not affect 
its market share, because my default browser is FF3.)

> the rate of adoption against other browsers. For example, Safari 2 to 
> Safari 3, or Firefox 1.5 to 2.0. Unfortunately, as you say, it is hard to 
> get reliable public statistics.

I'm not disputing that the adoption rate *inside* the IE browser family 
isn't that great. There may be many other reasons for it, such as IE7 
not being available for pre-XP systems.

But even with IE7 not being available (or usable) on older Windows 
machines, it still has the highest market share today (according to at 
least one statistic).

> However, that the second most widely used browser today is one that 
> shipped over half a decade ago, despite it being obsolete in its own 

Well, it stopped shipping only two years ago.

> product line _and_ having been marked for auto-upgrade to the latest 
> version by its manufacturer, is a pretty clear indication of what I am 
> talking about.

You said "...few people have adopted IE7 instead of IE6...", and I don't 
see how that follows either from the statistics, or from what you have said.

BR, Julian

Received on Saturday, 2 February 2008 21:18:34 UTC