next possible step
on my way to total Google-Graph-browsing nirvana: Have a search that uses the accumulated data (cache) and find out how many steps apart two given URLs might be. One problem I see here is that I can only fetch the top 10 related URLs for each URL, but maybe that is also a way of filtering... (ok, I _can_ fetch more than 10, but I really don't want to go into that. The automatic filtering these top 10 URLs give I'm rather happy with)