[Matthyze]

Thanks for the elaborate response!

I understand your interpretation, and I agree with the first part of it. (Don't change an existing system without first understanding the implications of that change.') I think that's the core of the metaphore, as taken by most people.

I don't think the point is that you should never make changes to complex systems at all, though. I don't think its means that more primitive, or unaltered, states of a system are necessarily prefential to more altered states, which I infer from your comment.

If unalterated states were better, we would have to tear Chesterton's fence down — right? Fences don't occur naturally.