I have been developing web sites since discovering Dreamweaver 2 in Year 9 Business Studies when I should have been learning about gross national product or something like that, which would start my web experience around 1995 I believe - with an additional year or two web browsing. I have been developing websites on a professional level for real clients (not just myself or friends) since I was at University in 2001 and I am now in my second web development job.

Back in '95, sites were designed using tables, HTML attribute style information and CAPITAL LETTERS FOR TAGS.

When Internet Explorer 6 came out it was a breath of fresh air, building on the past versions of IE and bringing support for CSS level 1, it opened up the door for keeping content and style separate.

In this modern day though, Internet Explorer 6 just doesn't match the demands from web technologies that modern designers and developers require for their sites. Plus, with Internet Explorer 7 being rolled out over Microsoft Windows Update, Firefox 3 updating 2 and Safari being available on platforms other than Mac OS, there is much more choice for the user to select for their web browsing duties.

When I'm building a website, I build it in my OSS environment of choice, and test it primarily in Firefox and Safari (since I am Mac based at the office). This is all well and good until this stage is finished and the site goes in to the dreaded stage of development - testing.

This is where I load up my virtual machine of Windows XP and check my site in IE6 and IE7. IE7 tends to be reasonably good in it's interpretation of the website code, only requiring the odd fix here and there. IE6 in the mean time has been known to render sites completely differently, mainly due to Microsoft and standards not fully agreeing on how big a pixel actually is.

It has occurred to me that I do spend an inordinate amount of development time fixing bugs in a browser that is slowly being phased out - and with the arrival of Internet Explorer 8 Release Candidate 1 sounding the charge of a sooner-rather-than-later final release - should developers be catering for this fading browser?

I was reading an article by Robert Nyman, a Swedish web developer, who feels similar to the way I do and has some rough numbers to support his article. I won't repeat them here, but you can read his article on his blog.