First scenario: say you have few hundred documentation files describing your software. Your new customer wants a copy of the files describing the new release you've just made. But those files all refer to the old version number and, worse, you've upgraded your company logo and some of them have the new logo, others have the old. Worse still your marketing people tell you that you should personalise this release by adding the customer's logo to the title page of each document, so they feel they own it. All your documents are in Word, or Open Office. What do you do? Oh, and they want them in PDF so you have to convert them all.
Another scenario: you write a document and email it out to others to review. They edit their ideas into the document and send it back. Soon you have fifteen copies of the document with different comments. The reviewers have not seen each other's comments so there is repetition (except two of them were emailing the same copy back and forth to each other, copying you, and you have four different copies from that exercise). Now you have to merge all the ideas together into something readable.
There are various solutions to the above scenarios, but maduradocs is a way to address these issues very simply using tools you already know, it runs everywhere, and it is free.
So far everything (including the source) is in a zip file. I'll put it into SVN sometime, but it is useable as is.