Jeb58,

I agree that the author's intent appears to be to indicate that defaults assumed in object name resolution aren't going to get you a cache hit, but there are all kinds of changes you can make to a query string that will lose the cache hit.  In fact, if I change anything in the query string, I lose the cache hit.  The presence or absence of an explicitly-named owner is just one way to mess this up, and since applications generate code the same way, I can get very, very good cache reuse from an application that never explicitly qualifies the owner.  This is the point that I think is - potentially - easily lost in translation in this article.

But you have your nickel, and I've only got my two cents.  If you understand cache reuse well, that's great.  But I'm suggesting this could be misleading to a newbie.  I certainly have my hands full trying to get some people in my organization to understand object ownership at all! 

Cheers,

Chris