The Opportunity Cost of Internet Explorer

Microsoft made headlines a few weeks ago when they asked customers not to use two of their products. One of the products in question is Internet Explorer, which was only included in Windows 10 for backwards compatibility reasons. The other is Office 2019, which is merely a cut-down version of Office 365 rather than a standalone offering.

Their new 'twins' marketing campaign for Office 365 got media attention for the way it denigrated the recently released Office 2019. The campaign features a series of video showing twins doing tasks in Microsoft Office. One twin uses Office 365 and the other uses Office 2019, with the former completing the task first. The aim is to highlight the feature differences between the two versions and drive customers towards taking out an Office 365 subscription instead of buying the traditional Office 2019 boxed product.

Subscription Models

The move towards subscription pricing for Office is still controversial in SMB circles but doesn't change very much for larger businesses who have preferred to pay for Microsoft products using subscription-based enterprise agreements anyway. Finance departments traditionally prefer subscription pricing because it appears as an Operating Expense rather than a Capital Expense on the balance sheet and allows greater flexibility in scaling license quantities up and down as business needs change. There will always be a contingent of software buyers opposed to subscription pricing either due to corporate culture or financial realities, but these are a minority. Changes in how recent versions of Office are licensed and installed have made life deliberately difficult for those businesses, leading to increased dissatisfaction amongst firms looking to purchase and manage software in the traditional manner.

Not all companies can move to the new modern management and cloud infrastructure paradigms favoured by the big enterprise technology vendors, either due to compliance concerns or the limitations imposed by the existing IT environment. There are tradeoffs in this new way of doing things that not all IT departments can cope with even if they wanted to. For many, the old ways still work fine and the evolving digital workplace hasn't impacted them enough to become disruptive. Stick in the mud IT managers definitely exist, but in many cases are concerned with trying to make things easier for change resistant users and investment shy bosses. Change is scary, and Microsoft have struggled to communicate the benefits to overworked IT staff that are too busy keeping the lights on to think about improvements to infrastructure. The success of Office 365 has been built upon the administrative benefits of having Microsoft host Exchange rather than doing it yourself. The fact you get the Office desktop apps and a bunch of other cloud services as part of the subscription is widely seen as secondary.

Technical Debt

Regardless of the human factors, the most significant barrier to IT transformation is generally existing infrastructure. This was best illustrated by the high amount of push back from IT managers to Microsoft's request not to use Internet Explorer. Very few IT managers want to use IE - they hate it as much as everybody else. Instead, its continued foothold in the enterprise is a result of compromises forced by existing devices and applications rather than a love for what Microsoft considers to be legacy technologies.

The typical enterprise uses thousands of applications. Some of them will be off-the-shelf commercial software, others will be relatively modern cloud services, but in every business there is the long tail of heavily customised or internally developed intranet apps that have been around since the mid 2000s, if not earlier. Most of these are browser-based apps developed according to the web coding standards of the time, which until 2010 were mostly whatever Internet Explorer happened to support at the time the application was designed. Standards have moved on significantly since then due to the rise of the mobile web and the additional capabilities required to support the trend of desktop level applications running in the web browser. As part of this, many legacy standards and development models – such as Flash or Java - have been removed from modern browsers, leaving Internet Explorer as the only one to support sites that require these deprecated technologies. Enterprise applications dating from the mid 2000s use these legacy development models because they were cutting edge at the time of release. The cost of rewriting and replacing all of these applications can be prohibitive, and in some cases can require extremely disruptive changes to business models or internal operations. It took a long time for the process to get going, but most IT departments are now making a concentrated to replace apps that only work in IE, but this will take time to complete, often years.

Long Term Consequences

This unfortunate situation facing businesses is a holdover from the era when Internet Explorer 6 held an effective monopoly in browser usage and Microsoft dominated desktop computing. Windows and IE still get hatred from developers over their attempts to push out competitors and redefine web standards around the turn of the millennium. They succeeded in these efforts, but then sat back and rested on their laurels, effectively stopping work on Internet Explorer for several years only to be surpassed in mindshare and usage initially by Firefox and then Chrome. Business decisions made by Microsoft 20 years ago still have practical day to day consequences for developers, who have been spent the past 15 years trying to get their projects to work in Microsoft browsers that don't support the latest standards. Few developers actively test in Internet Explorer these days except for sites where enterprise support for the long tail of businesses running Windows 7 is a requirement. When support for Windows 7 ends early next year, it is likely developers will abandon any pretence of supporting IE, especially because it hasn't been updated properly for 5 years. A lot has changed in web development during that time. Entire frameworks and UI paradigms such as react and flexbox have moved from bleeding edge into the mainstream. IE11 simply doesn’t support them properly, and the workarounds required to get them to work aren't worth the effort required.

Technology is constantly evolving and whilst a lot of effort is put into backwards compatibility, it is inevitable that as part of this evolution some things get left behind. Microsoft, in particular, have put a lot of effort into making Windows and IE backwards compatible with old applications, but there are limits to what can be done to enable legacy technology to work in modern environments. Technical advances are not just down to new features, they also result from new security requirements. Microsoft killed IE in the first place because the trade-offs needed to maintain backwards compatibility with 15 year intranet sites were preventing them from introducing new security models and modern web standards pioneered in other browsers and the wider community. These demands haven't gone away. If anything they've become stronger. Of course, there are downsides to this progress in technology too. Killing off an old app always results in disruption to the people who used it. The death of IE resulted in the loss of numerous features compared to its replacement and the migration of its userbase to Chrome. As such, Microsoft will be repeating the trick again later this year, replacing their Edge browser with a new version based on Google Chrome. There are trade-offs in that decision too, but if it makes Edge more useable than that's a net win for everybody. Just don't expect a mass migration from Chrome to Edge.