Add (past) discussion points to PEP 450.

pep-0450.txt

@@ -364,10 +364,48 @@ What Should Be The Name Of The Module?
     (described as "too C-like").
 
 
-Previous Discussions
+Discussion And Resolved Issues
 
     This proposal has been previously discussed here[21].
  
+    A number of design issues were resolved during the discussion on
+    Python-Ideas and the initial code review.  There was a lot of concern
+    about the addition of yet another ``sum`` function to the standard
+    library, see the FAQs below for more details.  In addition, the
+    initial implementation of ``sum`` suffered from some rounding issues
+    and other design problems when dealing with  Decimals.  Oscar
+    Benjamin's assistance in resolving this was invaluable.
+
+    Another issue was the handling of data in the form of iterators.  The
+    first implementation of variance silently swapped between a one- and
+    two-pass algorithm, depending on whether the data was in the form of
+    an iterator or sequence.  This proved to be a design mistake, as the
+    calculated variance could differ slightly depending on the algorithm
+    used, and ``variance`` etc. were changed to internally generate a list
+    and always use the more accurate two-pass implementation.
+
+    One controversial design involved the functions to calculate median,
+    which were implemented as attributes on the ``median`` callable, e.g.
+    ``median``, ``median.low``, ``median.high`` etc.  Although there is
+    at least one existing use of this style in the standard library, in
+    ``unittest.mock``, the code reviewers felt that this was too unusual
+    for the standard library.  Consequently, the design has been changed
+    to a more traditional design of separate functions with a pseudo-
+    namespace naming convention, ``median_low``, ``median_high``, etc.
+
+    Another issue that was of concern to code reviewers was the existence
+    of a function calculating the sample mode of continuous data, with
+    some people questioning the choice of algorithm, and whether it was
+    a sufficiently common need to be included.  So it was dropped from
+    the API, and ``mode`` now implements only the basic schoolbook
+    algorithm based on counting unique values.
+
+    Another significant point of discussion was calculating statistics of
+    timedelta objects.  Although the statistics module will not directly
+    support timedelta objects, it is possible to support this use-case by
+    converting them to numbers first using the ``timedelta.total_seconds``
+    method.
+
 
 Frequently Asked Questions
 
@@ -392,7 +430,7 @@ Frequently Asked Questions
 
        - math.fsum is high-precision, but coerces all arguments to float.
 
-       There is some interest in "fixing" one or the other of the existing
+       There was some interest in "fixing" one or the other of the existing
        sums. If this occurs before 3.4 feature-freeze, the decision to keep
        statistics.sum can be re-considered.