The connection to related content probably looks something like this:

This page that I tagged --> People who have also tagged this page -->  The tags that these people used for this page --> The pages that these people tagged using this universe of tags --> Rank by most frequently tagged

or perhaps like this:

This page that I tagged -->  My tags for this page -->  Other pages that other people have tagged using these same words    -->  Rank by most frequently tagged

This screenshot looks at the related pages to Mark Glaser's OJR story on "The good, bad and ugly of contextual ads from Google, Yahoo".  The related links are really interesting but not for the reasons that you'd think.  I would have guessed that a bunch of articles on AdSense and YPN would appear...some of which do in fact appear.  But what's interesting is that the results include links to some very thoughtful posts that are a little tangential to the specific topic, and yet the links are completely relevant to what I would want to know in this context.  For example, one of my links here is to the MarketingVox story "Study: Most Searchers Can't Tell Search Results from Ads, Use Search to Shop" and Poynter's "The 11 Layers of Citizen Journalism" by Steve Outing.

The unfinished part of our experiments in tagging back at InfoWorld were the connections to the world outside of our tags.  We showed related links based on the tags our own editors submitted with each of our own articles.  But the loop between our tags, other people's tags, and the people who also tagged our articles was not yet closed.  (Any progress on that front, Jon?) 

Joshua worked out an interesting relationship between the people who tag things and the tags that they share, a form of group knowledge sharing.  The first important milestone in this journey was getting so many people tagging successfully.  The ease of use of the tool and syndicatability (is that a word?) of the del.icio.us data opened up that possibility.  MyWeb then evolved the concept with the connection to your community and the filter that comes from that relationship.  Joshua then fired back with this feature which opens up a whole new world of discoverability.  What outcomes will people find from this new data?  What new tools and features will appear next?  I guess one question we need to start asking is, 'what do people want these tools to do?'