Why software engineering estimates are so often wrong

Courtesy of Marco.org, a link to an explanation -- an imaginary hike from San Francisco to Southern California.

I found this quite amusing, and there's a grain of truth to it -- no, actually there are several grains of truth to it. However, you rarely go into a development project as ignorant of the upcoming terrain as the imaginary hikers are. Still, the reality is that engineers often are inordinately optimistic about these things. Or if they're not optimistic, they're so pessimistic about the effort required that not only are their estimates no more useful than an optimist's, but their estimates antagonize their managers and can harm their careers.