IE6... usage not as high as we think?

I must ask the same question again: “What is it about IE6 that prevents you to design a great website?”

It’s not the design part that affects us, its the coding and the functionality with latest web standards that I hate about. We are in a time where the latest of stuff is the greatest. Sometimes some old things are just not meant to be. That’s how I feel towards IE6 . You’ll always have to do something extra just to make it function well.

I must ask the same question again: “What is it about IE6 that prevents you to design a great website?”

Nothing, but it does take a lot of time (and swearing) to make it work.

By the way, this thread has gone a bit off topic… i’m not complaining about supporting IE6 , i was just asking if 15% is a bit high because it includes those of us who HAVE to use it for testing purposes.

By the way, this thread has gone a bit off topic… i’m not complaining about supporting IE6 , i was just asking if 15% is a bit high because it includes those of us who HAVE to use it for testing purposes.

w3schools numbers are based on their server log files. When someone visits their site, the user agent is recorded and then this data is tallied to come up with the data they provide. What this means is that you using it to test your files on your server has no impact on w3schools’ browser stats. In fact as other people have mentioned, the numbers are likely to be skewed the other way as w3schools has a fairly tech savvy audience.

+1 Ezza

That’s one of the reasons I don’t trust W3C reports in the first place.

What about the Wikipedia browser usage reports?

Wikipedia is visited by a wide variety of users. I guess they’re using the same log files method, and this would be more accurate than W3C

I only found this – http://en.wikipedia.org/wiki/Usage_share_of_web_browsers

Maybe google has something… they’re the #1 visited site

Interesting ideas – and I suppose I never really realized how hot the debate is about these statistics. As with any un-scientific statistics though, I think you always have to allow for a 5-10% point fluctuation due to 1) tester bias, 2) insufficient or skewed testing, and 3) oddities such as X% of the actual IE6 users being web designers who are testing for IE6 . I don’t know enough to weigh in on whether or not testing is calculated in this group, but it seems like it’d be hard to filter out. My point about large corporations and other legacy institutions using IE6 as a standard remains though – this is not an insignificant audience.

I always just assumed that we’d all have to account for IE6 until those computers with IE6 loaded self-destruct or make their way to the recycling bin. It’s kinda like the era during the early 1900’s when automobiles and horse-drawn carriages co-existed… likewise, IE6 is one of those outdated tools of civilization that were helpful at the time, but now it just leaves horse$%#^ to clean up after while newer vehicles rule the highway.

Oh, and to the point of segen – why does IE6 matter when designing good sites. It doesn’t – like johnb41 said, it just stimulates demand for yet another level of testing, which can prove especially frustrating when you’re trying to implement a useful new trick that works on all other major browsers except IE6 .