This post originated from an RSS feed registered with Agile Buzz by Oliver Steele. Original Post: The Shadow of a Legacy Feed Title: Oliver Steele on Software Feed URL: http://feeds.feedburner.com/osteele Feed Description: Languages of the real and artificial. Latest Agile Buzz Posts Latest Agile Buzz Posts by Oliver Steele Latest Posts From Oliver Steele on Software

In software, an internally disruptive change is one where backwards compatibility is a major concern for the developing organization at the architectural level (and not just a matter of larger matrices in QA). It’s disruptive because it takes a drastic change to the architecture to cause this degree of disruption; it’s internally disruptive because, when compatibility is a requirement, the effects of the disruption are moved inside the development organization.

In the past twenty-five years, Microsoft has made taken Windows through internally disruptive transitions about four times; Apple has done so about thrice.

Now, this is a ten thousand foot view, and an approximation, and it’s not based on hard data. You could quibble: Maybe Windows NT should count. (I’ve counted its kernel changes against Windows XP, when they hit the consumer version of Windows.) Maybe MacOS HFS should.

But I’m interested in the shape of these transitions, not in their exact number. And the shape reveals the same thing over and over again: When Apple makes a disruptive change, it maintains compatibility back one version. When Microsoft makes a disruptive change, it preserves compatibility back to the beginning of (MS-DOS) time.

This is a difference in strategy, and its effect is cumulative over time. You can’t tell the difference from looking at the first internally disruptive change, but over time, the cumulative difference in strategies (indicated by the height of the bars) mounts up.

For a while, Apple was handicapping itself. It diverted its resources into OpenDoc, and into a series of next-generation operating systems (Taligent/Pink, Jaguar, Copland) that never saw the light of day. These diversions more than neutralized the cost of infinite compatibility, and Microsoft pulled ahead.

Now that Apple has been putting its development efforts back into one desktop operating system for a few years it has a widely acknowledged superior development velocity, and in many ways (in the consumer market) a superior position as well.

One look at the picture, and it’s easy to see why.