Why we’re stuck with IE6 for the forseeable future

If you’ve ever done any form of web development, you’ve probably learned to hate Internet Explorer 6. It’s not that it’s not used- IE6 once enjoyed the status of being the foremost browser on the web, and as a result used to set many of the standards by which the web was developed. Unfortunately, it never managed to consistently implement those standards, and as a result is the origin of many lost hours of sleep and endless frustration.

Internet Explorer 6 is old, very old. It was released on August 21, 2001, which in technology terms is archaic. To put things in perspective, the original, first generation iPod was released two months after IE6, shortly followed by the original XBox, subsequently followed by the first monochrome BlackBerry in March of the following year. Processors at the time had barely broken the 1GHz mark, and the Dot-Com bust had come and gone, leaving its mark on Redmond and Silicon Valley.

Since then, many newer and better browsers have been released, all of which do a better job of implementing those standards (Though few do so completely). As a result developing for these browsers has become even more painful, because web developers have to support both the newer, more standards compliant browsers as well as trying to accommodate for IE6’s eccentricities. And yet IE6 continues to appear on spec sheets and software requirements, and is a continued presence in web analytics reports. We’re still stuck with it, so what gives? If it is really so painful to develop for, and really so limiting to the user experience, why has IE6 not been unceremoniously ejected from the web?

Who’s Not At Fault

Blame is tossed around like candy when it comes to figuring out who’s at fault. One of the common ones is the outdated hardware, yet even in the notoriously underfunded and out-of-date nonprofit sector (this study claims 44% of nonprofits operate on 3+ year old hardware), the software is reasonably recent (89% on Windows XP). If anything, the nonprofit sector is ahead of the curve, perhaps because they have to rely on smaller software providers.

The larger part of the guilt is usually laid at the feet of large corporate IT departments. I’m not talking about mid-sized businesses, I’m talking about behemoths that, along with their size, have a reputation of moving incredibly slowly and always being several versions behind on everything. Some people erroneously say that corporations like this don’t want to shake things up by going with something “untested” and “potentially insecure”, but in reality the largest part of the blame for all the frustration they’ve caused isn’t even their fault.

One of the primary systems that large, corporate IT departments maintain are Enterprise Resource Planning (ERP) systems such as SAP, Oracle, Peoplesoft and so forth. These systems are massive- a full implementation will touch every part of an organization from accounting to inventory and requires several years to implement. Their benefits are many- they greatly improve reporting, planning, and in many cases automate core business operations.

Additionally, these systems are ridiculously expensive, not only because of their original licensing fees, but also because these systems have to be highly customized, and at $180/hour plus expenses for a competent ERP consultant that bill starts to add up. With a pricetag of many millions of dollars, installing an ERP may still be easily justified (given the operational benefits it would provide), however providing a positive ROI for the incremental benefits from an upgrade are a lot more difficult generate.

To illustrate, take a look at these two excerpts from a product compatibility matrix for SAP Netweaver ’04 and’07, valid as of April 2008 (the date on the presentation. You can find the version for ’04 here and for 7.0here , and as you can see they don’t exactly support anything but Internet Explorer. Add to that the extremely high cost and only incremental benefit of upgrading and it’s really no surprise that IE6 is still standard at major corporations. The ERP solution is the decision driver, not the browser, and that’s the real reason IE6 is not about to go away.


Product Compatibility Matrix for SAP Netweaver ’04


Product Compatibility Matrix for SAP Netweaver 7

So what happened?

Remember all that hooplah several years ago about how Microsoft was being uncompetitive and monopolistic in its actions within the browser wars? This is the aftermath. By continually encouraging corporations and their software developers to make use of the full capabilities of Internet Explorer 6 and the admittedly feature rich ActiveX controls, they have succeeded at carving themselves the largest part of the browser market. Unfortunately, by doing so they have also forced a large portion of the web into obsolescence.

So who’s really at fault here? Microsoft for pushing its development platform, or the ERP system providers who developed for it. Personally, I think the blame is where you prefer to put it based on your own technological preferences. If you’re a huge OSS fan, Microsoft’s a convenient target. If you’ve ever been frustrated by ERP’s, then the provider is the target-du-jour. Personally, I’m more likely to consider the product managers, directors and developers who made the platform decision without considering the long term implications.

So are we stuck?

Sortof. With the impending release of IE8 web developers will be able to tell the browser to “pretend” to be a previous version, neatly circumventing any future compatibility issues. Unfortunately this only works for Internet Explorer, meaning that ERP developers have no reason to change past behaviors. Until they disengage from using proprietary extensions and ActiveX controls and start relying more on open standards based development, we’ll never truly escape this cycle, though Microsoft has made backwards-compatibility a lot easier.

And that’s exactly what Microsoft wants: Allow their development partners to improve functionality and give their clients reasons to upgrade, while not forcing their other partners (IT departments dependent on ERP’s) into technological obsolescence… sortof. This process will take a while, because IE6 isn’t one of the browsers IE8 can be reset to, but eventually we will see IE6 go the way of Netscape, only to be replaced with… Internet Explorer.


  • Simon

    Hey! These pages don't scroll in IE6! What gives?

  • Oh, I know. It's just my own subtle rebellion against outdated browsers :)

  • Alec

    Great analysis. But at least we are able to drop IE 5.5. Surely you remember coding around that dog... IE6 can be coded into relatively graceful obsolescence. The visitors will see enough of the site that they can live with the broken edges.

Leave a Reply