Why is software development schedule estimation so hard?

As seems to be becoming a bit of a habit with me, I've decided to post an answer (with a few very tiny edits) which I left for a question that I found on the Programmers.SE website, mostly because I feel it's a pretty good answer, but also because it was an interesting enough question that I felt it would be "blog-worthy".

So the question was basically about two things, which was effectively to ask:

    Why is it so hard to provide accurate estimates for a Software Development Schedule, and 

    why isn't this as simple as planning a construction process, such as building a bridge.

Software schedule estimation isn't really very hard at all once you have some experience and can base your best guesses on tasks that appear to be similar to something you have done before. So if (as I'm suggesting) it isn't hard to estimate project schedules, why does it seem so difficult?