Reformulation of Day Count Bases

We have a few open issues related to day count conventions.  I went back 
to section 4.11.6 and found that it was still incomplete and rather 
difficult to reference cleanly from function definitions like YEARFRAC.

So I took the same material and tried to formula it in a clearer way (at 
least to me).  So I added a summary table at the start and then broke out 
each algorithm separately.  I also clarified that we actually have two 
sets of algorithms, one for day counting, the other for determining the 
length of years.  Finally I reformulated each algorithms as an explicit 
calculation.  I think this is less ambiguous than prose.

Please take a look and let me know if this is a preferable presentation of 
the basis conventions.  If so, I'd like to them verify that these 
definitions are actually correct by going back to David's earlier 
findings.

http://www.oasis-open.org/committees/download.php/39507/YEARFRAC.odt

-Rob