yep, agree. 
quick idea (without having tested it):

- schedule the httrack job (cron etc)
- write a script, which puts the resulting artifacts under version control
(git)
- keep the downloaded artifacts where they are (so don't delete or move them
after the git job)

then the procedure circles

- next time httrack job runs, it should (if applicable) overwrite the changed
files
- the subsequent git job will add it to the repository
- etc


by that you can keep all httrack runs separately (in a git repo).
if you like to see a specific version, check out that particular one from git
to a new folder and watch it in a browser.


the "index" file httrack generates won't get these links of course, but it's
just more or less a simple html, which you can create yourself and add the
starting index.html of each copy to it.
however, that would mean you would have to have the checked out repo versions,
which takes up (a bit) of additional space, since you will have so to say
duplicated copies (git repo + checked out version(s)).