1 files changed, 40 insertions, 0 deletions
diff --git a/doc/README.txt b/doc/README.txt
new file mode 100644
index 000000000..2a7ad82ee
--- /dev/null
+++ b/doc/README.txt
@@ -0,0 +1,40 @@
+NumPy Reference Guide
+=====================
+
+Instructions
+------------
+1. Optionally download an XML dump of the newest docstrings from the doc wiki
+   at ``/pydocweb/dump`` and save it as ``dump.xml``.
+2. Run ``make html`` or ``make dist``
+
+You can also run ``summarize.py`` to see which parts of the Numpy
+namespace are documented.
+
+
+TODO
+----
+
+* Numberless [*] footnotes cause LaTeX errors.
+
+* ``See also`` sections are still somehow broken even if some work.
+  The problem is that Sphinx searches like this::
+
+      'name'
+      'active_module.name'
+      'active_module.active_class.name'.
+  
+  Whereas, we would like to have this:
+
+      'name'
+      'active_module.name'
+      'parent_of_active_module.name'
+      'parent_of_parent_of_active_module.name'
+      ...
+      'numpy.name'
+
+  We can get one step upwards by always using 'numpy' as the active module.
+  It seems difficult to beat Sphinx to do what we want.
+  Do we need to change our docstring standard slightly, ie. allow only
+  leaving the 'numpy.' prefix away?
+
+* Link resolution doesn't work as intended... eg. `doc.ufunc`_