Good stuff, Elfwreck. I'm also thinking that the OP is starting to realise that the little rain drop they think they've stepped into is actually a huge, shark infested ocean. But either way, I do agree that there is a huge question of what the benefits of this system are vs the existing book sellers. Oddly though, you hit a large percentage of the questions right square on the head too. Those are all issues that would need to be addressed over the long haul before you'd have a viable product.
Now here's another kicker that the OP didn't consider. Payment processors. No matter what you do, you're still gonna have to pay the piper somewhere. So regardless if you do your own payment processing, or you go through google, Paypay, or someone else, you're still gonna pay someone something to process each transaction. The reason that you can get away with taking home 85% from places like Smashwords is because they have the benefit of raw volume mixed with low overhead. The average author doesn't have that.
Now, as for the OP, the best bet with your system will be to create a plug and play module that is drop in place, works with all the major CMS software systems (wordpress, xoops, jooma, etc. List of hundreds of systems available here =>
http://en.wikipedia.org/wiki/List_of...gement_systems), and has an API system for non-standard CMS software. (all my sites are custom coded from scratch) But when you look at just this tiny little aspect of the whole thing, you're basically trying to tame at tiger with a toothpick. Ultimately I see "epic fail" written across this project from the start.
Overall, given the hassle I already see with this, it'd be best to stick with what works rather than bother with reinventing the wheel. And this comes from someone who researched, and then ultimately abandoned the idea of selling ebooks directly from his site, and I'm a programmer, so projects like this normally don't scare me. This one however made me run for the hills.