I don't think a purge would get to the root of the problem, which is an incentive system that rewards adding early answers, rather than editing existing answers.
One of Joel's stated goals in the early days was that the answers would converge, Wikipedia-like, toward some Platonic ideal-- but the system in place doesn't encourage that.
The WP pages on Chrome or Git only mention major features of those pieces of software. Features don't change that much. Part of the articles talk about the historic aspects of the projects, and again the past doesn't really change.
SO tends to go into minute detail on a particular aspect of something, so the staleness effect is amplified. For example, questions on how to write a Chrome extension tend to be stale, as do ones on specific features of Git such as integration with Subversion, or submodule support.
One of Joel's stated goals in the early days was that the answers would converge, Wikipedia-like, toward some Platonic ideal-- but the system in place doesn't encourage that.