Are there any resources that explain the sort of URL normalisation Googlebot does?

Selected answers from the Dumb SEO Questions Facebook & G+ community.

• 
  Tony McCreath: Are there any resources that explain the sort of URL normalisation Googlebot does.
  
  Some normalisation is obvious, like ignoring case in domain names, but others are more interesting.
  
  Would Googlebot treat the following URL pages as the same address:
  
  /page?
  /page
  
  /page#
  /page
  
  /page?#
  /page
  
  /page?a=1&b=2
  /page?b=2&a=1
  
  /page?a=
  /page
  
  /category//page
  /category/page
  
  /category/../page
  /page
  
  /category/./page
  /category/page
  
  /page%2dname
  /page%2Dname
  
  I want to get a good handle on this before I delve into understanding how the GWT parameter tools works. So the main thing I'm interested in is parameter order normalisation.
• 
  Federico Sasso: Not sure about Google +Tony McCreath, but - for what it's worth - I certainly do all of them with VSS spider. I don't remember Google resources, I come to these conclusion mainly after dwelling into RFCs. Hope this helps.
• 
  Tony McCreath: +Federico Sasso ; I do some of them with my crawler. I was thinking of making it work as close to Googlebot as possible.
  
  The parameter tool and other Google activities imply they do play with URLs. In fact the parameter tool may be a clue into the mechanics they use. They are examining parameters so are they treating uniqueness based on parameter values and not just a querystring. And what happens if a website does not play by the rules:
  
  /page?aaaaaaaaa
  /page?=aaaaaaaaa
  /page?????aaaa????
  /page?======
• 
  Federico Sasso: Interesting.
  There is not a thing such a real "rule" when it comes to URL parameters: having querystring tokens describing name-value pairs is merely a common convention, but those are all valid URLs too (not sure about multiple question marks, actually). Google has certainly to deal with them too.