@@ -0,0 +1,44 @@
+.. _analysis-extractors:
+
+Analysis Extractors
+===================
+
+Using extractors and extractor-filters
+--------------------------------------
+
+.. note::
+
+    Extractor names are plural, even if they only extract a single item.
+    This is for two reasons. First, the common use of extractors is expected
+    to be to create an extractor-filter with the ``using`` method, in which
+    case it reads more clearly in the plural (since it will iterate over
+    many steps). Second, this allows a standard way to differentiate with
+    filters. For example, the filter ``canonical_mover(mover)`` gives a
+    filter that selects steps with the given mover/move type as the
+    canonical mover. On the other hand, ``canonical_movers(step)`` returns
+    the mover that is the canonical mover for that step.
+
+Built-in extractors
+-------------------
+
+* ``trial_samples``
+* ``active_samples``
+* ``shooting_points``
+* ``modified_shooting_points``
+* ``canonical_movers``
+
+* ``canonical_details``
+
+Writing custom extractors
+-------------------------
+
+Extractors API
+--------------
+
+Extractor-Filters API
+---------------------
+
+In general, the way to create and use an extractor-filter is to create it
+with the ``extractor.using(step_filter)`` method, and then to wrap the input
+``steps`` with the resulting extractor-filter. Users will rarely directly
+initialize an extractor-filter via its ``__init__`` method.
@@ -0,0 +1,52 @@
+.. _analysis-filters:
+
+Analysis Filters
+================
+
+For a detailed overview of the using extractors and filters to simplify
+analysis functions, see ???. This page will document built-in filters, as
+well as the underlying API for creating new filters.
+
+
+Built-in filters
+----------------
+
+Step filters
+~~~~~~~~~~~~
+
+* ``all_steps``
+* ``accepted_steps``
+* ``rejected_steps``
+* ``shooting_steps``
+
+* ``canonical_mover``
+* ``trial_replica``
+* ``trial_ensemble``
+
+
+Sample filters
+~~~~~~~~~~~~~~
+
+Frequently, you'll extract a list of samples from the step, e.g., using
+either the ``active_samples`` extractor or the ``trial_samples`` extractor.
+In this case, you may want to select only specific samples (perhaps based on
+the ensemble) for your analysis. Sample filters make this possible, and can
+be used as the ``secondary_filter`` when creating extractor-filters.
+
+* ``all_samples``
+
+* ``ensemble``
+* ``replica``
+* ``sampling_ensemble``
+* ``minus_ensemble``
+* ``ms_outer_ensemble`
+
+
+Writing custom filters
+----------------------
+
+Filters API
+-----------
+
+
+