I like the reference to creating a site about refrigerators, but the
only thing I could write about refrigerators is something on the order
of how to try to figure out what something that got lost in the back of
the bottom shelf was when it was still somewhat identifiable.

Is there anyone who thinks that this Supplemental Index business has
little or nothing to do with crawling efficiency and how decisions are
made which sites to crawl, when and how much? This is an old paper, but
it still makes a darn good read

Efficient Crawling Through URL Ordering

http://dbpubs.stanford.edu:8090/pub/showDoc.Fulltext?lang=en&doc=1998...

If we're dealing with what geek types are doing, then we need insight
into how geeks think.