IBMZ-Linux-OSS-Python
diff --git a/‎doc/library-developers.rst‎
Lines changed: 5 additions & 4 deletions b/‎doc/library-developers.rst‎
Lines changed: 5 additions & 4 deletions
diff --git a/‎patsy/splines.py‎
Lines changed: 8 additions & 5 deletions b/‎patsy/splines.py‎
Lines changed: 8 additions & 5 deletions
@@ -16,9 +16,10 @@ Patsy is a good houseguest:
 * Pure Python, no compilation necessary.
 * Exhaustive tests (>98% statement coverage at time of writing) and
   documentation (you're looking at it).
-* No dependencies besides numpy (and we even test against numpy 1.2.1,
-  as distributed by RHEL 5).
-* Tested and supported on every version of Python since 2.4.
+* No dependencies besides numpy.
+* Tested and supported on every version of Python since 2.5. (And 2.4
+  probably still works too if you really want it, it's just become too
+  hard to keep a working 2.4 environment on the test server.)
 
 So you can be pretty confident that adding a dependency on Patsy
 won't create much hassle for your users.
@@ -162,7 +163,7 @@ it's quite straightforward to add custom operators to the formula
 parser, or otherwise extend the formula evaluation machinery. (Heck,
 it only took an hour or two to repurpose it for a totally different
 purpose, parsing linear constraints.)  But extending Patsy in a
-more fundamental way this will require just a wee bit more complicated
+more fundamental way then this will require just a wee bit more complicated
 API than just calling :func:`dmatrices`, and for this initial release,
 we've been busy enough getting the basics working that we haven't yet
 taken the time to pin down a public extension API we can support.
 
@@ -103,11 +103,14 @@ class BS(object):
     A spline with ``degree=0`` is piecewise constant with breakpoints at each
     knot, and the default knot positions are quantiles of the input. So if you
     find yourself in the situation of wanting to quantize a continuous
-    variable into equal-sized bins with a constant effect across each bin, you
-    can use ``bs(x, num_bins, degree=0)``.
-
-    Similarly, a spline with ``degree=1`` is piecewise linear with breakpoints
-    at each knot.
+    variable into ``num_bins`` equal-sized bins with a constant effect across
+    each bin, you can use ``bs(x, num_bins - 1, degree=0)``. (The ``- 1`` is
+    because one degree of freedom will be taken by the intercept;
+    alternatively, you could leave the intercept term out of your model and
+    use ``bs(x, num_bins, degree=0, include_intercept=True)``.
+
+    A spline with ``degree=1`` is piecewise linear with breakpoints at each
+    knot.
 
     The default is ``degree=3``, which gives a cubic b-spline.