Ok, well maybe the example provided doesn't communicate the intention, because when I try to 'extract' the page, following the download links, they don't work, except the png. Cool concept though.
Unfortunately, the file poses a problem for basic unzipping software (e.g. a file explorer). However, the file is 100% valid with respect to the ZIP specification. You just need to use "true" unzipping software like 7zip, unzip etc. to read it as such.