> The site I am trying to spider uses relative links (eg. a
> href=".". a href="..", a href="../.."), but HTTrack 
doesn't
> seem to deal with them properly.  Instead of translating
> them to the correct absolute url it creates a file called
> ..html and makes a link to that.
> Have I set it up wrong

Yes: httrack will always stays in the same directory (or 
lower directory structures), by default (except external 
images fetched by default with scan rules)

Use 'Set options'/'Scan rules' to override this behaviour ; 
using for example:
+www.example.com/*

or even more customizable:
+www.example.com/myfiles/*
+www.example.com/foo/*
+www.example.com/bar/foobar/*