A fresh look at Internet Explorer support
Google recently declared that their web applications will no longer support Internet Explorer 7, a decision I welcomed publicly on Twitter when the announcement was made. Because Internet Explorer has historically treated web standards with a rather cavalier disdain, supporting its various versions can add significantly to any website’s development cost.
But what does Google’s decision mean for web developers? Should we withdraw support for IE7 when creating websites? After all, it’s only a year or two since we were all debating whether to retain support for IE6!
It seems to me that there are at least two factors at play here – firstly whether users are able to upgrade or change their browser, given their operating system; and secondly, whether they are willing to.
Can’t Upgrade…
For instance, people running Windows XP cannot upgrade to Internet Explorer 9 – it simply does not work on that operating system. Here’s a closer look:
| Compatibility | IE6 | IE7 | IE8 | IE9 |
|---|---|---|---|---|
| Windows 2000 | ||||
| Windows XP | ||||
| Windows Vista | ||||
| Windows 7 |
This table suggests that only people running a version of Windows that is older than XP would be unable to run a website if support for Internet Explorer 7 is withdrawn. (We can ignore the Mac platform for now – Internet Explorer hasn’t run on Mac since version 5.)
But do such people still exist? What are the numbers?
Well, Wikipedia tells us that Windows 2000 has 0.2% market penetration. In my book that’s small enough to disregard, particularly when you consider that these die-hard users also have various other choices available to them, including decent versions of the Mozilla Firefox and Opera browsers.
So, enough users can upgrade to Internet Explorer 8 that this particular factor should not persuade us to retain support for Internet Explorer 7.
…Won’t Upgrade
So much for old versions of Windows. What about actual usage of Internet Explorer 7?
| Usage | IE6 | IE7 | IE8 | IE9 | ALL IE |
|---|---|---|---|---|---|
| May 2011 | 10.36% | 7.04% | 31.28% | 4.19% | 52.87% |
Wow. So more than 17% of internet users are still using Internet Explorer 6 or 7. This starts to make Google’s decision look a little braver.
The higher-than-expected figures for Internet Explorer 6 have always been explained by commentators to be corporate users whose choice is dictated by custom web applications that may only work on IE6. So if your website isn’t designed to be used by that audience while they are at work, ignore them. Don’t forget they probably use a better browser when they’re at home.
However, the Internet Explorer 7 figures suggest that a large number of people can upgrade to Internet Explorer 8 without changing their operating system, but have not done so. When it comes down to it, these are the people we have to consider when deciding whether to retain support for IE7.
Full or Partial Support
Seeing as about one and a half billion people (roughly a quarter of the global population) have internet access, the 7% still using Internet Explorer 7 actually represents quite a few people. Quite possibly a larger number than you would want to alienate with a completely broken website.
However, there are broken websites, and there are websites which maybe don’t look perfect but still let you get to the information you need. And I think this is the key to the issue of browser support.
Often, by spending €X, you can build a website which looks perfect on IE8 (and Firefox, and Safari, and Chrome, and Opera), and works but doesn’t look so great on IE7. Making it look perfect on IE7 may push the cost up by 25%. That extra 25% is being spent on around 7% of users, who in my view are probably not the kind of user who cares whether something looks perfect. After all, they don’t care enough about their computing experience to upgrade their browser.
Biting the Bullet
About a year ago, I stopped trying to make my clients’ websites look perfect in IE6 (unless they paid extra, to cover the additional development time). Functionality was still provided, but if the layout broke in some way, no matter.
And although I’m not quite as brave as Google, I believe that now is a good time to do the same for IE7 – and to drop even basic functional support for IE6.
| Tweet |
|
18 Jun 2011
It’s a bit funny where most geeks in the business will just shrug their shoulders and say “too bad” if a customer refuses to upgrade a program like office or quick books and the customer is unwilling to do it.
But of old version of IE we will spend hours and even days trying to get a element to move over 3 pixels to look the same in IE as in all the browsers that work with the standards.
MS has a history of ignoring standards hoping to set “the standard” themselves. All this ever leads to is buggy competing software and web design that helps nobody in the end.
An interesting article, but it depends very, very heavily on which part of the world you’re looking in. Check out China – http://gs.statcounter.com/#browser_version-CN-monthly-201104-201106 IE6 is still the dominant browser (40%), closely followed by IE8 (36%), so if you have any interest in attractive Chinese students, investment, whatever, then you absolutely *must* support IE6 (although for some reason it seems that you can ignore IE7 – perhaps two different populations like government institutions / universities on IE6 and personal machines who have been able to upgrade to IE8 like the rest of the world).
But look at the USA – http://gs.statcounter.com/#browser_version-US-monthly-201104-201106 – and it’s a very different story – Safari on the iPad outstrips IE6, so IE6 is effectively dead. IE7 is hovering around 6%, which really seems like we could start ignoring it now – particularly as broadband penetration is so high and there’s not much cost to downloading big upgrade files.
The amount of additional hours I’ve had to spend on websites because they don’t display properly on the older versions of IE is unreal. However – most of those still using old versions of IE aren’t doing so by choice. It’s usually down to the fact that businesses and establishments (libraries for example) are often reluctant to upgrade their systems. I think they need to be better informed as to the security risks that come with using an old browser.
It’s sadly still the case that you need to design for all the browsers that adhere to industry standards first (firefox, opera, chrome and yes – even safari).
Then see what it looks like on ie(7,8 and now 9), and add new css for each.
I honestly wish MS would drop this browser – or at least only make 1 version per year as it’s a complete pain…
I think they don’t upgrade because they fear to learn new things. It seem using upgraded software is hard for a lot of people
I actually work for a company that has a very important online application that requires only IE 7. Its crazy that they would use that and not update their code to support newer browsers with much better features. Great article, I have bookmarked your site. Perhaps we could do a link exchange my main page has a PR of 4. Thanks!
Installed on Windows 7, and my experience is:
- It doesn’t install IE9 (?).
- The machine already had IE8 installed, so it didn’t install that.
- Can’t get IE7 on Win7, yeah, ok, I get that.
- So it installed two versions of IE6, neither of which work, they just crash on launch. I tried running them as Administrator, no luck.
So I started out with a machine with IE8. And now I have a machine that still has IE8, but also two broken installs of IE6.
Im surprised nobody has mentioned solutions the like of Chromeframe. I now make a Chromeframe popup appear for anyone using IE6 or lower, and a ‘recommendation’ for Chromeframe appear for those using IE7.