-There is really no play button right now.The outer keys seem to jump to the end and the inner keys move one step, but there is no play button really.  There should be adjustable playback speeds or something to effect to control the temporal unfolding.

-A bar indicating where the time is in the current playback would be nice as well. The time bar could be interactive to an extent; one could place a comment on the time bar so when the playback reaches this point a speech bubble indicating a certain event (a preposition for example) in the diagramatic unfolding could pop up for a specified period of time.

-The record button should stand out  while recording, it should change colors to indicate recording in progress. At this point I don't really think the record does much beside allow drawing. A record session should last one on and off period of the record button pressing. The record session should be stored or prompted to store after it is ended, or some means of keeping track of various changes or different projects. Maybe an alternate window containing a file directory. Somewhat like the sentence window on the stemma program, but with a directory of record sessions.

-It would be nice if one of the keys on the wacom table restricted the pen mode of the tablet to exist within the boundaries of the drawing screen. This would be 'tablet mode' or something.

For the future:

-When objects are eventually implemented, you could avoid object recognition in the program by making a menu of items that are manipulable, such as a circle that can distort into an oval, a square with adjustable size, a barrier with a movable gate etc. to satisfy the current diagramming quest. Object recognition would eventually be a nice feature.

-In addition to object recognition, in the far future, maybe we could try to see if patrizia could help us implement a speech recognition aspect. The object menu could be executed by speech: "circle" "barrier" etc, with hopes to eventually make a more complex system that can be used in classroom settings to aid in creating diagrams. (For more information on this embodied learning environment see: http://cognitiverobotics.pbwiki.com/Embodied+Learning+Environment )

-Thinking even further ahead, maybe this program could eventually be used to recognize speech, parse it, and reconstruct the thought in a diagrammatic form. The system could be trained by extracting patterns from diagrams that are manually created and annotated either with voice or sentence descriptions that will be parsed and all of the input (drawing, typing, speaking) entered in a database along with the stemmatic realizations to create a more robust worldly knowledge in which patterns, such as those created in the plotline formation of temporal stemmatics, would be used to model the narrative of the drawing in order to extract patterns.