As for the rest of it - is it possible to 'dump' the html from the Wiki?
Not really. Internally, all of the pages are stored as Wiki code, and are only converted to HTML when a page is requested. I do think that including an archive of the Wiki in the official docs is great idea though, so I'm currently writing a little script to covert the Wikicode into Text/HTML/PDF. I'll post a link here or on the Wiki when it's ready.
We could 'prime' the Wiki with the current official docs and people can go nuts with it from there.
That sounds like a good idea, too. As long as you don't have a problem with the docs being placed under the FDL, we can just copy and paste everything to the Wiki to help get things going.