We have a capsule which does some kind-of complicated processing such as writing to a text file from a dataflow and reading it back in to some work cubes and then displaying a screen based on the work cubes. If more than one user is using the capsule it works alright. When user A loads the work cubes and displays results and user B then loads the work cubes and displays results then user A's view of the data is not affected unless they refresh their screen and then they see unexpectedly user B's version of the data.
I am wondering if the data could be separated by adding a user ID entity as a work cube dimension, loading it be referring to the substitution formula @user (or @username) and filtering the work cube that way. The hoped-for effect is user A would have work cube dimensioned with A and user B would have work cube dimensioned with B and regardless of whether someone refreshes their screen they would only see their own data. The capsule will potentially be used by several users at a time.
Any thoughts, work-arounds, suggestions?