If you know anything about web design, then you will know that one of the major bugbears of any web designer is coping with the multitude of browsers that proliferate on t’internet. There are 4 or 5 major players, including Microsoft Internet Explorer, Mozilla Firefox (our favourite here at Four Lakes HQ), Safari (the de faco Mac option), Opera (with a small but oh so determined following) and the new kid on the block, Google Chrome. If you cover all these then you will cater for most of your web site’s likely visitors, or at least those that are not using a smartphone. But that’s a whole other kettle of fish which we might address in the near future.
The issue with that list, though, is that there isn’t necessarily a single browser representing each player. Go have a quick look at an excellent resource for web developers, BrowserShots. On here you will find a wider listing (although by no means totally comprehensive) of the available browsers, down to the different versions of each one on different OS’s. Putting in a website URL, you can choose in which browsers you want your site rendered. Taking Firefox on windows as an example, you could display your site in 0.9, 1.0, 1.5, 2.0, 3.0, 3.1, 3.5 and 3.6. Wow!
You might think that there is potential for a hell of a lot of work, making sure your site works as intended across the multitude of browsers out there. In fact, it’s not so bad. One thing that most of these browsers have in common is that they follow the prevailing standards (in HTML, CSS, etc) at the time they were released. Of course, the standards evolve over time, adding new functionality to support such goodies as Ajax, etc. This is one reason why you get major revisions of browsers rather than a gradual evolution. Therefore, if you create a site that utilised standards-compliant code, then you can be pretty sure that it will render properly, at least in the later versions of the major browsers.
In the past, though, there has been one exception to the standards-compatible path, and you may not be surprised to learn that it’s Microsoft’s baby, Internet Explorer. Microsoft has a history of, shall we say, promoting new standards, and in version 6 of Internet Explorer, released in 2001, there is much that treads a different path to its contemporaries. Delve deep into the code of any half-decent website and you are likely to find whole sections devoted to addressing IE6-specific rendering issues. For web developers, it’s an outright pain in the posterior!
But why bother, I hear you cry? Surely IE6 has been superseded by IE7 and now IE8? Well, actually, not as much as you might think. It’s a bit of a long story, but in essence it boils down to the fact that many large organisations, commercial and otherwise, have not upgraded their PC’s from XP to Vista. I know, it sounds a bit far-fetched, but hear me out.
Large organisations like to have standard ‘builds’ of the operating system and a set of supported applications for their PCs. They are then secured so that end users cannot easily add, change or remove applications. By doing this they achieve a couple of advantages. Firstly, if the PC breaks they just replace it with another PC with a standard ‘build’, and the user is back up and running in minutes rather than days. Secondly, they can test the standard application set to death to make sure that all the needed services work properly. Therefore if, for instance, the company uses Salesforce.com, they can verify that it works properly in their chosen browser. Making a change to that standard build involves a fairly intense and expensive exercise of test, approve, build-update, documentation-update and roll-out.
Of course, when XP was released the pre-installed browser was IE6. Companies, in general, were happy to migrate from their previous OS to XP (with IE6) and go through the extensive testing process to ensure everything worked. There was an incentive, then, for websites to be IE6-compatible or risk disenfranchising a huge section of their potential audience. If Vista (with IE7) had been a compelling choice for large installations, then another round of testing would have occurred and, no doubt, IE6 would have faded out. But the general opinion in the corporate world was that Vista wasn’t good enough to warrant the expense. So even now, despite the greater success of Windows 7 and the proliferation of alternative browsers, IE6 is still used by about 20% of people.
That’s why I was particularly interested in an email I recently received from Google about the support for IE6 in Google Apps. Google have announced that, from March, anyone accessing Google Apps using IE6 will be politely informed that their 9 year old browser is out of date and they should consider installing a newer one. They obviously feel that the risk of disenfranchising some of the 20%, and perhaps causing some of the larger companies still using XP/IE6 to finally go through the build update process, is now worthwhile. The benefit to Google is that a lot of IE6-specific development work will no longer be required, and perhaps some functionality that could not be made to work in IE6 can finally be introduced, although the latter is pure speculation on my part.
I for one am looking forward to a brave new world that doesn’t include IE6. Here’s hoping lots of other companies follow Google’s lead on this.