Monday, June 7, 2010

OpenMessenger First Week

So yeah. Not much to say about OpenMessenger. Last week I went through all of the sensors and monitors attached to the program to make sure that they work. Here's a rundown:
  • Keyboard Monitor: Works. Needs a longer timer between each input process.
  • Microphone Monitor: Works. Needs a new threshold for actually reporting activity, but the differences between real noise and background noise is significant.
  • EyeTracker: Unknown. I will be setting up and calibrating the EyeTracker this afternoon with Lindsay.
  • Screen Sensor: Works. Needs a monitor written to actually process the data.
  • Webcam Sensor: Same as above.
In general, we are looking into alternate ways for the Monitors to report a lack of activity. Currently they are broadcasting a lot of events when nothing is going on. If we code the OmniWindow to only show changes in activity and change the monitors to only send events on changes, we can cut down on network traffic a bit as well as only display information that is relevant.

Speaking of the OmniWindow, there are a few things that need to be added, including Mouse support for debugging purposes, as well as code to display activity icons. The activity icons should fade out over time to make sure we keep clutter to a minimum.

This will be my last blog post for two weeks as I'm going on vacation this Thursday for 10 days. Matt Lepage will be working on OpenMessenger in the meantime, and for the rest of the summer. We hope to have OM working by the end of the month, and be able to devise a study using it around then.

1 comment:

  1. Glad to hear things are working better than planned Johnathan. Enjoy the time off!

    ReplyDelete