Kerika uses a ton of Javascript, and by the word “ton”, we mean “well over a hundred thousand lines of Javascript”. It is one of the most sophisticated user interfaces ever developed for the browser, and it delivers a fantastic, real-time, desktop-like experience right inside the browser. And it does so on Internet Explorer, Safari, Chrome and Firefox.

The problem has always been with Internet Explorer: versions before IE9 have very poor support for Javascript and HTML5 in general. In years past Microsoft gave lip-service to the idea of supporting the (then-emerging) HTML5 standards, but their pre-IE9 versions did a very halfhearted job of supporting Web standards. Too many people at Microsoft were still fixated on the idea of maintaining a proprietary lock-in by encouraging their users to stick with the Microsoft-specific extensions that ran only on IE Server and the IE browsers. Things turned around only with IE9, spurred, no doubt, by the surprising inroads that were being made by Firefox and Chrome. Now, happily, the thinking at Microsoft has really come around to supporting Web standards!

Well, Kerika doesn’t run on pre-IE9 versions of Internet Explorer. This was a critical design decision we made when we first started coding in 2010, based on the assumption that IE9 — which was then in beta form but already looking reasonably robust — would be quickly launched and vigorously promoted by Microsoft, and as quickly adopted by the IE user base.

That doesn’t seem to have happened, at least not as far and as widely as we had hoped, and a call this afternoon with a user who was stumped trying to make Kerika work on his office PC, which still runs IE8, prompted us to look at NetApplications data on browser market share and adoption curves. The data are, frankly, dismal.

First of call, let’s look at overall browser market share, as of Dec 2012:

Overall, Microsoft is in a position to say they are still the most common browser out there, but not by much: total market share for Internet Explorer, across all versions, is 54.77%. The following graph, however, really puzzles us: it suggests that browser market share hare remained essentially static for most of 2012, which doesn’t quite jive with anecdotal evidence we have been getting from users suggesting that Chrome is making surprising inroads among both Mac and Windows users:

Looking at specific versions of Internet Explorer shows some data that matches our intuition and general market understanding:

But there’s other data that are really puzzling: why would IE6’s market share rise in mid-2012? Were a bunch of old laptops suddenly taken out of storage and donated? No new machines could have come into the market with IE6, nor could many machines have downgraded to IE6 (unless everyone has been saving their installation disks for Windows XP and decided to collectively reinstall their desktop operating systems in July…)

Perhaps the data aren’t so reliable after all, although NetApplications has long been the most highly cited source for data on browser market share.

We need some of those curves to start bending soon…