Microsoft announced this month that they will be using Windows Update to upgrade Internet Explorer 6 & 7 users to the latest release, IE8. Although Microsoft insist you will still have to opt-in to the upgrade, it will be set as ‘High Priority’ for many users, which effectively means that if you don’t want it you’ll have to opt-out. For the online world, this is a pretty big milestone in browser history: the end of IE6.

IE6 was launched on the 27th of August 2001. Back then, and only if you were up to date, you would have been using Windows 2000, a dial up modem and all your applications would have run on the desktop. Amazon had yet to make a profit, newspapers still thought people would pay to access their sites, Google wasn’t the dominant search engine, and GMail and Facebook were still three years in the future.

I think it’s fair to say that the 21st century web was built with IE6 as a dominant browser. For many years it was the dominant browser, and even now retains enough popularity in enough sectors to still be considered a major browser – it still has a roughly 17% market share as of March 2009. Not bad for a 7½ year old browser.

So what’s wrong with it?

Well, for a start, the browser is old. For a browser, old means less secure, it means not standards-compliant, it means it was built for a simpler, more innocent online world and today, that means broken. When support is finally withdrawn, IE6 will be nearly as old now as the web itself was when it was launched. It really doesn’t support standards properly, instead offering a broken partial support that I’m convinced has caused more stress, swearing and overtime than any other piece of software in history. Nowadays you can develop a site for any standards-compliant browser, be it Firefox, Opera, IE7, Safari in one go, and then have to start a new development process just to fix it for IE6, with all the costs that entails. Any other browser so broken would have been ignored long ago, but as part of Windows XP and still accounting for nearly 1 in 5 web users, we can’t. Yet.

So who is still using it?

The browser is still popular (well, common) in enterprise environments, places usually with thousands of centrally administered machines running standardised desktop configurations. Rolling out a change to these standard desktops is fraught with risk – a missed software conflict can cause downtime in the hundreds of thousands of hours, data loss, damaged customer relations and countless shredded nerves before the problem can be fixed. You don’t change these things lightly.

In the eyes of many IT departments, the browser has also changed from a dangerous source of viruses that must be controlled and locked down, to a business critical tool servicing dependent software that must be kept secure and running at all times. As a result, upgrades were and are few and far between, or usually nonexistent. If it works, why fix it?

Progressive enhancement to the rescue.

The problem with maintaining a legacy browser comes when you need to develop new online applications, and, as will be the case with IE6 in July 2010, support for your browser ends during the lifetime of the project. Many due diligence rules and guidelines demand that all software, especially business critical software, be fully supported by a vendor. The approach to this transition many companies have followed is to require all new projects work on new browsers and legacy ones. This isn’t a problem in itself of course, but when the request is for the project to not only work across browsers, but work in exactly the same way, and look the same, it becomes a problem.

Progressive enhancement is all about developing a functional, usable and attractive basic experience that can be delivered to all users and that will allow them to do what they need to do. To this you add features, aesthetic improvements, faster responses, friendlier layouts, improved usability and so on, but only as far the user’s browser can display them properly. The idea is that your site doesn’t prevent anyone from visiting and becoming your customer, but people with better browsers get a progressively better experience. One important thing to note is that all of this happens automatically; the user is not required to choose between ‘have’ and ‘have not’ varieties of your site, nor should they ever be aware that anything is ‘missing’.

For people continuing to use IE6, however, it may mean a drop in the quality of their online experience, as one by one development teams class it as a ‘basic’ browser and no longer spend extra hours making enhancements work with it. Interestingly, a major factor driving this change will be the global recession. Providing a ‘full experience’ to IE6 will increasingly become a chargeable item as developers find they can no longer absorb the significant costs it incurs, and companies that use and commission online applications will be under increasing pressure to keep those costs down, and upgrade.

For everyone else, we can expect a higher quality experience – without all that time spent making everything work in IE6 we can make things work better for standards compliant browsers and clear the way for the next generation of web applications.