Right now, if an experiment is in piloting mode it won't save data. This used to be fine, but it's now causing some issues with session experiments (see #41). Thus, I'd like to propose a change in how we handle piloting data.
As we now always have a session object which tracks the state field of the experiment when it was created, we could use that to seperate actual data from piloting data. Thus, I think we should now actually save piloting data as well. This would mean we need to make the following changes:
Change the upload views to save piloting data as well
Change the experiment detail view to display piloting and actualy data seperatly (using tabs maybe?)
Change the download view and code to handle downloading both types of data seperatly
Change the group assignment code to only look at sessions that have the same state as the experiment when assinging groups
Otherwise, piloting data would impact the balancing and quota of actual data
(In jsPsych utils) change the message that data won't be saved, as it now will be
Right now, if an experiment is in piloting mode it won't save data. This used to be fine, but it's now causing some issues with session experiments (see #41). Thus, I'd like to propose a change in how we handle piloting data.
As we now always have a session object which tracks the state field of the experiment when it was created, we could use that to seperate actual data from piloting data. Thus, I think we should now actually save piloting data as well. This would mean we need to make the following changes: