Yes, good questions. I anticipate that I will initially log a fair amount of data while the app is running, logging the data in a "batch mode." That is, I will log data for a number of minutes (10 - 20) and then retrieve that data and analyze it. I suspect that the quantity of data I log will decrease with time as I gain confidence in the app and fix errors.
When analyzing the data I expect that some of the tabular, numeric data might be plotted, possibly against other "truth" data using engineering tools such as Matlab. (For example, I might plot the filtered acceleration data recorded by the app vs the raw acceleration data sent to the iPhone app.) There might also be debug data consisting of time stamped text strings with an associated severity eg, info, warning, error, fatal. (For example, I might generate a log entry with the text version of verbal feedback produced by the app.)
So, yes, I will be processing the collected data, largely for development but then a smaller subset for ongoing verification of the app.
Thanks.