From: ben.usenet@bsb.me.uk   
      
   Tim Rentsch  writes:   
      
   > Ben Bacarisse  writes:   
   >   
   > [...]   
   >   
   >> I found this problem interesting but only later in the discussion as   
   >> I have been using this metric for some time.  What got interesting   
   >> (to me) was that there is another sound interpretation of the   
   >> average as suggested by Tim, [...]   
   >   
   > It wasn't my idea.  I got it from a posting by Mike Terry on   
   > December 21.  I hadn't seen that formulation of arithmetic mean   
   > before and I was amazed that it worked.  So I can't really take   
   > any credit for the suggestion.   
   >   
   >> ironically prompted but a general definition of what might   
   >> constitute an average that I had posted and failed to follow   
   >> through on.   
   >   
   > I remember your posting as coming after the one by Mike Terry,   
   > and so I thought your comments were derived from his.  Sorry if   
   > my conclusions there were off the mark.   
      
   You are right in that MT posted the same general formulation 9 hours   
   before I did (though I'd not seen that).  My confusion came from your   
   explanation, to me, of "conventional average":   
      
     "Sorry, I meant to refer to your formulation of average"   
      
   followed by the formula I gave rather than then entirely equivalent one   
   given by MT.   
      
   Anyway, the credit I'm giving is for your considering this a reasonable   
   thing to try calculate for arc lengths, rather than the vector average   
   that minimises a different measure altogether.  Maybe MT also suggested   
   that as an alternative but I only remember his championing the vector   
   interpretation.  My apologies to MT if he did that as well.   
      
   --   
   Ben.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)