I presented the model of mobile collaboration I defined (derived from the CatchBob results as well as coordination theories) to my PhD advisor. It actually addresses the exchange of various kinds of interfaces to foster coordination among a mobile group of players. There are actually two tracks to validate the model described previously:

A tool that would foster the exchange of coordination keys, to better support collaboration. In the form of a structured interface, this tool will suggest the exchange of certain kind of strategy messages (instead of automating which failed as we saw in the first experiment). The analysis of the interface usage will allow us to validate or refine the model by checking when specific keys are exchanged during over time.
A formal description (in the form of a grammar) of coordination elements that would help the analysis of mobile collaboration. Using along with the replay tool, this grammar will help characterizing visually how users collaborated with regards to peculiar processes: coordination keys exchange, division of labor, duration of subtasks… The validation of the model will consist in using this grammar with the replay tool to the differences for groups who badly collaborated or for those who collaborated efficiently. We already know what are the “good” and “bad” collaborators (related to the task performance and various indexes), we will see whether the grammar fits into that picture. In the end, this grammar is meant to allow a better comprehension of collaborative processes in mobile teams.

The research process, visually speaking: