Forum Moderators: open
I just read something from 2006 where the author was saying he had dropped IE5 and IE5.5 support for his product. And that got me thinking about the recent hoopla over Google trying to force upgrades from IE6 to non-IE browsers.
So I wondered if anyone still supports and tests IE5 and IE5.5. How many people have given up testing in IE6? I long ago gave up testing in IE5-5.5, but in truth I pretty much don't even check IE6 anymore. I used to have all the standalones running, but now on rare occasions just go to a machine with IE6.
I checked out browser stats from the last 11,000 and some odd visits on one site. Here's the breakdown among IE browsers. Since overall IE is about 52% of visitors on this site, you can divide by two for overall numbers. So in other words, only about 18% are still on browsers <= IE6. That's almost one in 5, but I'm surprised by how much it has dropped since say a year ago.
7.0 -- 7,147 - 63.45%
6.0 -- 4,067 - 36.11%
8.0 ----- 30 - 0.27%
5.x ----- 14 - 0.12% (5.0, 5.01, 5.14, 5.16, 5.17)
5.5 ------ 6 - 0.05%
Your IE6 numbers are a LOT higher than ours, ours have dropped to around 17%. And our IE8 is running around 2%.
Once IE8 is out in final, and no major problems arise, I expect to pretty much stop tweaking most things for IE6.
[edited by: Wlauzon at 10:51 pm (utc) on Jan. 8, 2009]
That said, at the time I took the decision all my sites worked perfectly in 5.5 (and, I would add, in IE5 for Mac, though not 5.01 for Windows). As far as I am aware (I just checked one) they still do - but I tend to avoid any technique that causes compatibility problems and prefer to keep my sites relatively simple.
I certainly don't lose sleep over IE5x and haven't met a Windows 98 user for ages.
...
I don't do IE < 6 anymore, as a matter of fact: I don't have any of them to run them anymore (XP comes with IE6 already, and I never want to go back to running multiple IE versions side-by-side, virtual machines rule).
I can't afford to stop supporting IE6 yet, but it's becoming the typical rock in my shoe that soon or later will have to go ...
IE8's general availability might be the trigger, dunno, will see how fast it gets adopted.
On some less important websites I might stop supporting IE6 sooner, and redirect those still on IE6 to a "please upgrade your browser" page (a conditional comment should do the trick nicely). I'm not sure I'll point them to IE8, might be a browser with a bit more tradition to support standards ;-).
I'm not letting MSFT off the hook for their IE8 blocker tool that will work indefinitely.
Msie 8.0: 28 - 0%
Msie 7.0: 368407 - 63.7%
Msie 6.0: 63298 - 10.9%
Msie 5.5: 9 - 0%
Msie 5.23: 11 - 0%
Msie 5.00: 4 - 0%
Msie 5.0: 26 0%
So, 430,000 hits from IE7 and IE6, compared to 11 for IE5 Mac and 39 for IE5 PC (and they're probably bots too). So IE5.x is totally gone for me on all my sites, I certainly don't bother testing for it any more.
IE6 share has dropped dramatically since IE7 was pushed through Windows Update, although obviously at 10.9% it is still far too high for me to consider dropping support, but I could consider some graceful degradation of the experience if required.
The following is pure guesswork, but I expect the IE6 figure to drop to under 5% before the end of 2009. I suspect also that it will still be a significantly-significant number for a couple more years, in particular for sites aimed at corporate audiences.
And although some corps will still be stuck on IE6, the vast majority of our actual buyers are individuals.
And "not supporting" IE6 does not mean that all of a suddent things won't work on our sites. It does mean that we are not going to spend much - if any - time tweaking for one browser.
Are you any less aggressive about testing in IE6 than you were a year ago? Two years ago? If I go back to the summer of 2007, 40% were still on IE6 on the same site as seen above (again, sample of about 12,000 visits).
significant number for a couple more years, in particular for sites aimed at corporate audiences.
I should also have mentioned that another inspiration for the thread was a discussion yesterday with my wife who is one of the few people in her company who is allowed to browse with anything other than IE6. Out of a few thousand employees with work computers, I think there are fewer than 10 who have IE7.
Universities and public libraries are other places where there is great resistance to change.
18% is a pretty big number to kiss off.
Well, I wouldn't say kiss off. I would say that even if I'm just building a new page on an existing template, I usually view it in both IE7 and FF, but I would not in that case check it in IE6. IE6 is most decidely an afterthought. And now that I think of it, I don't even have a machine with IE5 set up on it any more. So now I have two windows open when I tweak (FF3, IE7), whereas a few years ago it was more like five or six (IE5, 5.5, 6, FF2, O).
In my numbers Safari and Chrome account for almost as many as IE6. It's probably more important to test Safari than IE6 at this point. Opera, on the other hand, with 0.60% is tested only because doing so is easy, but increasingly rare for me.
>>resolutely non-technical and connecting from home
Likely the same with my figures.