

placeholder: pics of robot
Recap: The main goals for this project was to capture \interactions that are unique to the new social culture that is being created from web communications. Many times, we find ourselves bringing high-level social gestures to the system-level abstractions. Whether it be linking through webs of friends, "writing on walls", or skyping with a friend studying abroad, a key element prevents this interaction from being interpreted as "real", but rather is given a class of its own.
The project breaks that class by bringing concrete physical interaction from the virtual space. A robot was constructed from the repurposed remains of a toy remote car, and Arduino that handled electronic functionality with remote signals, a laptop that hosted a Java server that relayed OSC (open sound control) signal from a MAX/MSP/Jitter patch and a Flash front-end. The flash site was uploaded to a web server. The robot could thus be controlled through any computer connected to the the same network router.* A live webcam feed was connected to the robot and streamed through means of the media server on Justin.tv.
The laptop display also handled OSC receiving of messages from the user.
The current flash interface can be found at:
http://stanford.edu/~ctorres7/TEST.html
The current interface currently contains an instructional, however from preliminary tests with the robot, this will be removed. In order to bypass network control and see the full interface of the robot (unconnected) hit CTRL+ENTER on the homescreen. A button will be placed for easier access.
The robot was set up at Stanford and was allowed given a test drive. Almost immediately, we found that a question of a virtual person or body within the confines of a photography free zone is constituted as a person. Many bugs and further functionality need to still be worked on, but I hope to further explore this area with this art piece.
SKIP TO 2:15.
* Consideration were made for world-wide access to the robot controls, however this required port-forwarding and MAC cloning of university internet configurations that we protected by multiple firewalls. Testing will be done to reconfigure a more accessible network connection to provide full access. Perhaps even a network card from a phone 3G service..
So, last week I realized I did not like the direction this project was going. I felt that much of the interactivity capable by the MAX/MSP/Jitter language was being limited by computer vision problems and inaccuracies. Instead, after seeing ACCESS, "an interactive installation that lets web users track anonymous individuals in public places, by pursuing them with a robotic spotlight and acoustic beam system" by Marie Sester, I decided to do a redesign of the project.
I particularly still wanted to keep this concept of virtual space and idenitity (two places at once) and the idea of interacting physically from a distance. This project will deal with constructing a "robot" from a toy remote car which will be receiving OSC messages from a server locally hosted on my computer. The robot will have a camera for guidance and an LED ticker that can be used to display messages.
I feel this project has a lot of potential due to the anonymity of the beholder and the forced interaction of passerbys. It really test the bounds between the 2 dimensionality of social web applications to a less human exterior, but far more human interior of a human-guided robot.
Goals for this week:
Redesign UI and get FlashServer running for video feeds.
Design robot. I am thinking to make it out of cardboard similar to the architectural topographic models (the ones that look like level sets) to get past the robotic stereotype (chrome and lights).
Outer columns and singular. Inner columns are merged.