Microsoft has a pretty strong reputation for maintaining backwards compatibility for applications that run on their platform. They have to, since the continued operation of many thousands of critical applications depends on them doing so. Microsoft understands this special position and responsibility, and has put an amazing amount of work into keeping old program working. Just study up on the base Win32 C API to see what I’m talking about.
What’s surprising, then, is the debacle of the Internet Explorer 7 release. Surely, with IE 6 having such a large market share, it would not be difficult to realize that there many web apps out there that depend on the platform provided by the IE6 HTML and Javascript runtime. Yet, even under such obvious circumstances, Microsoft released IE 7 which broke many a website, even forcing some organizations to avoid it all together.
This blunder seems to indicate a few things. Microsoft still doesn’t really understand the significance of web apps, nor their existing presence in this area of technology. They could have released a significantly enhanced IE7 that didn’t break IE6 sites, at least maintaining their market share, and perhaps allowing websites and organizations to jump on new IE7 features without hesitance. Instead, they made everyone realize the importance of web standards and probably also made a few people take a good look at Firefox (both are good things).
For a company that had historically done pretty well at releasing mostly strict improvements to their previous products and, most importantly, maintaining backwards compatibility for applications, IE 7 (and Vista, to also to a large extent) seem like a great departure from their winning strategy. Maybe this is on purpose, but it’s hard to see how this has worked to their benefit. It seems counter to the strategy of taking advantage of their market dominance and installed base.
Then again, it is also very technically difficult to keep winning the backwards compatibility game. I’d say this is especially true for things like Javascript, where MS doesn’t have complete platform control (they define Win32, where as Javascript grows more organically). Maybe their engineers finally gave up and decided to work on something new. Maybe IE7 and Vista are the results of this. But it’ll be interesting to see what happens as these products arrive when there are already comparable market equivalents (Firefox and OS X). Breaking their support for legacy apps puts their competitors on a much more level playing field.