Second Iteration

  
Figure: Second iteration opening screen. The amount of screen space given to each part of the interface is determined by the status of the user. This screen is the experimenter's screen.

In order to more accurately reflect the responsibilities of the respective users, two separate opening screens were created. The screen that a user worked from would be determined by the login status of the user as described above. Figure  illustrates the experimenter's screen, while figure  illustrates the operator's screen. Here, the path which the user would typically travel occupies the majority of the screen space while the less used path is represented by the smaller box. The user would then activate part of the screen with the mouse and continue along the task hierarchy.

The task hierarchy that the user traverses during an experiment was determined by a generalization of all the actions required to complete an experimental run, data acquisition, and on-line analysis. These tasks were divided into the following five categories:

• detector setup: the series of steps required to prepare the detector hardware for data collection
• experiment setup: includes hardware set up and crystal preparation
• beamline setup: includes beamline calibration and alignment
• data acquisition: collection of data once a data run has begun
• experiment status: steps taken to verify experiment is proceeding well

Distinction between the two user groups within this categorization was achieved by the allowance of access to certain functions. Parameters which were not open to experimenters for manipulation were ``greyed out". Characterizing the distinction between operator tasks and experimenter tasks within this categorization proved problematic. At the same time, maintaining similar screens for each type of user without acknowledging the differences in user task performance caused many redundancies in screen display and access to functionality. This resulted in a generic interface, devoid of any clear path through the tasks, because functions could be accessed from different screens. For example, a novice user would not know if, a distance parameter was part of equipment setup or detector setup. The latter would imply the need for intervention by an operator.

  
Figure: Second iteration opening screen. This screen is the operator's screen.

Later evaluation resulted in a recategorization of the tasks mentioned above. Description of this would be futile, for the categorization of the experiment syntax proved redundant and arbitrary. Clearly, the solution to the problem did not lie in this area.