Well you spot anything new, and it's not the first time I read something like this. As you notice, until the market share of IE6 drops under % we should always take care of it when producing web contents. Obviously I'm referring to sites that actually include that kind of target. Currently I'm working with this key in mind: for mainstream browsers the site should be "perfect", on IE6 it should be "usable", never spending too much time on adjusting margins or strange absolute boxes positions: at least the user will not be disappointed and can browse through the site without problems.

Hi,
I happen to work in a big company (300.000 employees) and guess what the company browser is? Right: IE6. For years and years nothing has changed here and only because so many people are using Firefox for surfing they company slowly sees that there are new browsers and browser versions on the market.

The problem is always, that a lot of custom-written, very old applications that happen to be business critical rely on IE6 or an outdated version of Excel or Access. So the change cost and risk when migrating to a newer version or even another browser is quite high.
So don't expect the big businesses to change their old ways anytime soon...

The problem is not with the size of the company, but rather, with the minds & lifestyles of the people!

In my company where about 300 workers are present, all the Intranet apps work with IE6. And it's still not the 300 workers that matters, it's just 'ONE' application developer who plays only within the MS-Environment, and whose mind works in only one direction: Writing code in .NET Framework.

The managers of the company backs up the MS Environment under this argument:
Microsoft is the best! Which other platform could be a better option? Nobody knows.

When opposed and say the Firefox should also be supported, says our MS- developer: Hah, the apps will be run only within the company Intranet. Don't waste your time trying to make it other way.

The basic and minimalistic app and Web UI design should be the solution.

You check your logs, and you see what percentage of your actual users are still coming in with IE6, and then you make the call. I have been known to develop sites which present the same content with a slightly different appearance for IE6 users, but it's more of a passive-aggressive tendency than a technical issue. IE6 users don't deserve smooth-blended transparent drop-shadows. :)

At this point the IE6 issues are so familiar, I find myself writing the hacks into the CSS before I even test. I'm always more concerned about making a site which degrades easily to semantic HTML for less capable browsers. With that as a basis, it is so much easier to adapt your CSS to support legacy browsers.

My vote went for, "No, but we probably should". The truth is that I test rendering of all sites I design in both IE7 and Firefox, and occasionally in Safari...but honestly, if it looks good in Firefox it will usually render correctly in Safari as well.

While it's a good point to judge your design effort time put into making it look right for a certain browser based on user stats, the point is moot if you are developing for a new domain and entirely new web site. So I just always check the following sites (you'll have to Google the sites because I don't want my comment filtered)

BrowsrCamp - Safari screen capture of web pages

IE Net Renderer - Displays your page in IE7/8 (beta) or IE6 and also shows a combo rendering of 6/7 or 7/8 if you like, which applies an overlay of colored areas that show any difference between two versions. Only shows about the top third of the site, though (above the fold)