How about instead of a built-in page refresh, find a good RSS aggregator and subscribe to the Newest Nodes feed. Then you can set in your aggregator exactly how often you want to ping for new nodes along with all the other typical aggregator functions. (If you do this, set your refresh on the node to something odd like 42 minutes or 71 minutes or something.)
For additional PerlMonks RSS feeds, check out this page.
You may also want to consider using these feeds as source data for a Perl script using LWP and XML::Simple. Then you'd have the power to do just about anything you can image with the data. Here at work, we have a Perl script that pings for newest nodes every so often and posts the title and URL in an IRC channel on an internal server to which most of our developers subscribe.
Whitepages.com Development Manager (DSMS)
code('Perl') || die;