Fix-up deprecations language.

1 files changed, 13 insertions, 5 deletions

diff --git a/doc/source/reference/c-api.deprecations.rst b/doc/source/reference/c-api.deprecations.rst
index 4a5184ca7..86f15527d 100644
--- a/doc/source/reference/c-api.deprecations.rst
+++ b/doc/source/reference/c-api.deprecations.rst
@@ -7,11 +7,14 @@ Background
 The API exposed by NumPy for third-party extensions has grown over
 years of releases, and has allowed programmers to directly access
 NumPy functionality from C. This API can be best described as
-"organic".  The core API originated with Numeric in 1995 and there are
+"organic".   It has emerged from multiple competing desires and from
+multiple points of view over the years, strongly influenced by the
+desire to make it easy for users to move to NumPy from Numeric and
+Numarray.   The core API originated with Numeric in 1995 and there are
 patterns such as the heavy use of macros written to mimic Python's
 C-API as well as account for compiler technology of the late 90's.
 There is only a small group of volunteers who have had very little
-time to spend on improving this API.   
+time to spend on improving this API.    
 
 There is an ongoing effort to improve the API to do things like remove
 a "fortran" macro and ensure that the NPY_ prefixes don't collide with
@@ -35,9 +38,14 @@ Deprecation Mechanism NPY_NO_DEPRECATED_API
 -------------------------------------------
 
 In C, there is no equivalent to the deprecation warnings that Python
-supports. One way to do deprecations is to flag them in the documentation
-and release notes, then remove or change the deprecated features in a
-future version (NumPy 2.0 and beyond).  
+supports. One way to do deprecations is to flag them in the
+documentation and release notes, then remove or change the deprecated
+features in a future major version (NumPy 2.0 and beyond).  Minor
+versions of NumPy should not have major C-API changes, however, that
+prevent code that worked on a previous minor release.  For example, we
+will do our best to ensure that code that compiled and worked on NumPy
+1.4 should continue to work on NumPy 1.7 (but perhaps with compiler
+warnings). 
 
 To use the NPY_NO_DEPRECATED_API mechanism, you need to #define it to
 the target API version of NumPy before #including any NumPy headers.