diff --git a/IPython/zmq/taskthread.py b/IPython/zmq/parallel/taskthread.py
similarity index 98%
rename from IPython/zmq/taskthread.py
rename to IPython/zmq/parallel/taskthread.py
index b1824ea..bddaa23 100644
--- a/IPython/zmq/taskthread.py
+++ b/IPython/zmq/parallel/taskthread.py
@@ -12,7 +12,7 @@ except:
 import zmq
 from zmq.core.poll import _poll as poll
 from zmq.devices import ThreadDevice
-from IPython.zmq import streamsession as ss
+from IPython.zmq.parallel import streamsession as ss
 
 
 class QueueStream(object):
diff --git a/docs/source/development/parallel_messages.txt b/docs/source/development/parallel_messages.txt
index 7755c4f..3ece970 100644
--- a/docs/source/development/parallel_messages.txt
+++ b/docs/source/development/parallel_messages.txt
@@ -63,8 +63,6 @@ Message type: ``registration_reply``::
         'control' : 'tcp://...', # addr for control queue
         'heartbeat' : (a,b), # tuple containing two interfaces needed for heartbeat
         'task' : 'tcp://...', # addr for task queue, or None if no task queue running
-        # if error:
-        'reason' : 'queue_id already registered'
     }
 
 Clients use the same socket as engines to start their connections. Connection requests
@@ -172,8 +170,6 @@ Message type: ``result_reply``::
                          # values are the result messages
         'pending' : ['msg_id','...'], # msg_ids still pending
         'completed' : ['msg_id','...'], # list of completed msg_ids
-        # if error:
-        'reason' : "explanation"
     }
 
 For memory management purposes, Clients can also instruct the controller to forget the
@@ -199,9 +195,6 @@ Message type: ``purge_reply``::
 
     content = {
         'status' : 'ok', # or 'error'
-
-        # if error:
-        'reason' : "KeyError: no such msg_id 'whoda'"
     }
 
 :func:`apply` and :func:`apply_bound`
@@ -210,7 +203,7 @@ Message type: ``purge_reply``::
 The `Namespace <http://gist.github.com/483294>`_ model suggests that execution be able to
 use the model::
 
-    client.apply(f, *args, **kwargs)
+    ns.apply(f, *args, **kwargs)
     
 which takes `f`, a function in the user's namespace, and executes ``f(*args, **kwargs)``
 on a remote engine, returning the result (or, for non-blocking, information facilitating
@@ -220,8 +213,8 @@ as little data as we can. The `buffers` property of a Message was introduced for
 purpose.
 
 Utility method :func:`build_apply_message` in :mod:`IPython.zmq.streamsession` wraps a
-function signature and builds the correct buffer format for minimal data copying (exactly
-zero copies of numpy array data).
+function signature and builds a sendable buffer format for minimal data copying (exactly
+zero copies of numpy array data or buffers or large strings).
 
 Message type: ``apply_request``::
 
@@ -234,6 +227,8 @@ Message type: ``apply_request``::
     buffers = ['...'] # at least 3 in length
                     # as built by build_apply_message(f,args,kwargs)
 
+after/follow represent task dependencies
+
 Message type: ``apply_reply``::
 
     content = {
diff --git a/examples/demo/dagdeps.py b/examples/demo/dag/dagdeps.py
similarity index 100%
rename from examples/demo/dagdeps.py
rename to examples/demo/dag/dagdeps.py
diff --git a/examples/demo/loadbalance.py b/examples/demo/loadbalance.py
deleted file mode 100644
index f87015e..0000000
--- a/examples/demo/loadbalance.py
+++ /dev/null
@@ -1,20 +0,0 @@
-import time
-from IPython.zmq.parallel.client import *
-
-def wait(t):
-    import time
-    time.sleep(t)
-    return t
-
-client = Client('tcp://127.0.0.1:10101')
-view = client[None]
-
-tic = time.time()
-for i in range(128):
-    view.apply(wait, 1e-2*i)
-    # limit to 1k msgs/s
-    time.sleep(1e-2)
-
-client.barrier()
-toc = time.time()
-print toc-tic