Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

  • Improved consistency accross dialogs: based on a somewhat misguided guideline, we had implemented some actions of our dialogs using hyperlinks (e.g., 'Cancel'), and uthers using buttons (e.g., 'Save'). Our users complained, and we got rid of the links and adopted buttons for all dialog actions.
  • Combined similar actions: our users complained that the interface had similar actions (e.g., 'Create assignment' vs. 'Create multiple assignments') and that added some confusion, so we unified them ('Add asignments').
  • Use attribute names consistent with the real world: in Rails, it is common to name timestamp attributes ending with the suffix '_at". This surfaced to the interface and the evalulators requsted it to be changed to 'Due at'. A similar problem happened with our implementation of a 'Class' object; because 'class' is a reserved word in Ruby, we named our structure 'Course', and it surfaced to the UI.
  • Added information about Haag on the login screen: some evaluators were concerned that it was not possible to know what the application does, so we added a short explanation on the landing page.
  • Implemented missing features: we implemented the missing features from GR4, most of which were picked by the evaluators: allow unsibscribing from a class; implement undo for marking assignment as completed; allow creating private assignments.

Implementation

Our application was implemented using the Rails framework (version 3.1.3). For effects and widgets, we used the jQuery UI libraryon the desktop version, and jQuery Mobile on the mobile version.

Rails applications are structured according to a model-view-controller (MVC) pattern, and are expected to adhere to a series of guidelines. This "convention over configuration" philosophy benefits the designer by ensuring that if the guidelines are followed the low-level details of the application will be handled automatically. This allowed us to have the first features up and running in a short time.

Evaluation

We found three undergraduates for the user tests. One has a calendar with all of her assignments for the entire year, one makes a list for each upcoming week, and one just keeps everything in his head. We managed to test someone who matched each of our original user scenarios.

...