I spent today working on the data which we retrieve from the app, namely by adding a lot of options to how we interpret that data. A lot of it was behind the scenes infrastructure, such as editing the .plist files from xcode to reflect our study, then creating a new file to house a plain text dictionary of all the texts. I thought about trying to link the .plists to that text document so that an edit in xcode would be automatically registered in the text dictionary, but couldn’t find a good way of doing it, besides having all editing be done through the text dictionary which wouldn’t be half bad I suppose). From there I made a new python program and class called Data Analyzer, which essentially takes in the data from Vidushi’s program and my own and tries to pull as much data as possible out, such as word length vs. acceleration, longest word in a subsequence vs acceleration, and tons of other data can be added. I did this in an attempt to emulate Rayner’s work, but as I said yesterday, I think trying to work with saccades and other purely eye focused metrics may not be suitable for our study simply due to the irritation caused by the quickly scrolling text, particularly at large text sizes. I plan tomorrow to try to recreate Vidushi’s program in my new environment so that everything is self contained, as well as go through and significantly clean up my code/comment thoroughly so others may use it without hassle. Also it could do with a bit of optimization so I’ll also tackle that tomorrow.