What I’d like to do at some point is to build a fulltext Atom feed that uses Feed Thread and Feed History extension elements. You could then read Perlmonks from right inside your aggregator just like on the site, threading and all. (Using Feed History would offset the cost of a fulltext feed by ensuring that aggregators don’t need to poll aggressively to avoid missing feed entries.)

(Since they’re not even completely finished and blessed, there isn’t much support yet for these extensions – but that would be helped if there were incentives for it in the form of existing consumable data. Which would in turn be a happy side-effect of having a clean static archive of Perlmonks, which IMO is worthwhile all by itself. I always worry a bit about the fact that the years of community knowledge and history that have accumulated at Perlmonks are stored in what’s effectively a closed silo. Be nice to have a copy of it outside the database in a future-proof standard format.)

Makeshifts last the longest.