For the first time in roughly a year, I saw what my sie looked like in IE6, and I was distraught. Not a single page renders the way it should, with some practically unusable. With IE8 on the horizon, is it worth it to remake my CSS to work with IE6?

As of right now, we're talking about roughly a third of our visitors are IE6, a third are IE7, and a third are Firefox. However, the IE6 statistic is decreasing by the month. While 30% is a rather significant audience to cut off, I'm just wondering whether you spend the extra time and effort to design for older versions of IE?

yeah i think no browser perfect,
but i think firefox id better right now

I've been reading up on it, and it turns out that IE6 has a LOT of CSS bugs that were corrected in IE7. Unfortunately, it still holds a 30% market share, so it looks like developers are being forced into supporting it.

By looks of it you may need to support IE6 at least for another year...

Yep... IE6 users don't like to upgrade, simple as that and we will need to keep supporting it. Since I am not a fanatic for valid style sheets (see no valid reason to be) I am thankful that IE6 display bugs are relatively simple to fix using underscore (importing extra stylesheets are just too much work to be worth the trouble IMO).

Matti Ressler

Yes my first site some how did not work on ie5-6 and some times it would mess up with opera, and sea monkey/firefox but it work perfect on safari and ie7

I work at a web development firm in Boston. All the designers and programmers around me are Firefox users. The firm I am at does a lot of "big" client projects. Anyway, one of the partners put a sign on the desk of the guy that sits next to me with an IE6 logo speaking "Please don't forget about me or the millions of people that use me."

We test our sites EXTENSIVELY in IE6, FireFox, IE7. The partners are ALWAYS bothering us about testing in alternate browsers. I big into Microsoft, so they are always joking with me: "When did you last test in Firefox? Does it work in Firefox?"

In short, you NEED to test in IE6!!!!!

I think if we make website for clients, we must test it in different browser.
I usually use ie6 and firefox to test it until it work well in that 2 browser.

I think part of the problem with IE7 is that its such a drastic departure from what traditional IE6 users have been used to for such a long time (in terms of appearance etc). Now with IE8 being in testing, we should (hopefully :icon_smile: ) see lots more IE7 and IE8 users, or perhaps converts to Firefox.

Just a matter of time!

Your present user figures speak for themselves. You certainly need to keep supporting IE6 until it's in negligable use.

The best way to solve the problem is to---from the beginning---only use techniques that are known to work everywhere, where 'everywhere' currently, unfortunately, includes IE6.

Since you've just done a bit of a redesign and it looks good from where I'm sitting ( Opera ); if I were you I wouldn't invest too much time making everything look perfect in IE6. Just get the functional part working ( you said the site is 'practically unusable' in places, so sort that bit out ), but leave any tiny visual discrepancies alone, or assign them a very low priority, and then just watch the user browser figures. Maybe in 6 months things will be different. Maybe in 6 months all those IE6 sufferers will find Opera ^_-

You know, Opera users may well appear as IE6 users. Personally, I always indentify as IE because certain sites refuse to open up to browsers that identify as Opera, so.. maybe your user stats are a little skewed by Opera users ( for which the site works fine ). Last time I looked at the browser ID string in this mode ( was Opera 8, sometime last year ) it was equiv. to the IE6 browser ID string.

ie7 is a high priority update in microsoft update. This is probably why the numbers are going down. I am not sure but I think aol still uses ie6 as their primary browser. old joke:How many Microsoft programmers does it take to screw in a light bulb? None. Let's define darkness as the new industry standard.

As a general rule, you typically design for Firefox, since it is the most standards compliant browser out there, and then fix for IE6, IE7, and Safari as needed. FF has a ton of awesome developer tools (Firebug, etc) to get the job done, so that's another reason to start with it.

Your numbers certainly indicate that IE6 is an issue, and I imagine it will be for at least another year. Imagine if you lost 30% of your revenue? I tend to make it at least usable in IE6, and if it's a little rough around the edges, maybe it will remind them to upgrade.

As a general rule, you typically design for Firefox.

I agree - this is exactly how I develop. Get it right in Firefox first, then make corrections for other browsers where necessary.

For IE6 I am not afraid to used underscored classes in my stylesheet, since this is the quickest and simplest way to make corrections for it. Purists insist on importing stylesheets, however IMO this is unnecessary overkill with no gain whatsoever other than more development time.

My only qualms with Firefox are its massive memory leaks.

Matti Ressler

Unfortunatly people still use IE6 even though with IE8 and Firefox 3.0 coming out it will take a little longer to make sure the Website is perfect in all browsers but until IE6 is down to like 10 percent I would keep supporting it :(

Hey Dani, I don't know if you saw the statistics by the w3 of how many people use what browser so I will post them as of april 2008 IE7 24.9 percent IE6 28.9 percent IE5 1.0 FX (firefox) 39.1 percent Moz (netscape) 1.0 percent Safari 2.2 percent Opera 1.4 percent
here is my source

A good coder shall support IE6 and FX2 and never bother with any other; should I add: browsing toys, which are more seriously concerned with their own look rather than with how they render your reqiured content.