We are currently using random query-strings to prevent Internet Explorer from caching HTML on postbacks, but we are in the process of implementing URL re-writing and it would be nice if we could remove these random parameters.
What’s the best pratice in this situation?
Have you tried the no-cache and expires meta-tags?
Other than that, random generated query-strings are the way to go. In my opinion, its not not that big of a deal. domain.com/stories/man-walks-on-moon/s34b is not exactly making my eyes bleed. And you will still achieve great indexing as far as search engines go.