Part four of a series on the conservation of Max Dean’s As Yet Untitled. Get up to speed on the project.
By Sherry Phillips, conservator of Contemporary and Inuit Art
This week, we sent off As Yet Untitled to VOX, Centre de l’image contemporaine, where it will be included in Le Mois de la Photo à Montreal, running Sept. 5 to Oct. 5, 2013. It was a hectic week leading up to packing the installation, and every day that we ran the robot program we learned something new: usually something quirky, possibly undesirable, but something that had to be addressed nevertheless.
There is a certain amount of imprecision in programming and teaching the robot. Much of the process is trial and error and repetition and running out to the local electronics store for supplies (see below). We spent long hours simply turning the control unit off and on, repeating robot actions by manually moving the arm back to its zero point, testing alignment marks, and watching and waiting for discrepancies, anomalies and tics. Once those arose, we began the long process of looking for and correcting the potential source of the problem, which could be as simple as a misplaced period within a line of code.
As Yet Untitled is time-based media and performance art. We needed to patiently teach it how to move through its performance by establishing the coordinates for each of its five joints: the shoulder, arm, elbow and two wrists. To get a sense of how it moves, check out another famous Canadian robot, the Canadarm 2, which has an impressive seven joints in all.
I eavesdropped with interest and admiration to the telephone chats between Marcel and Richard, and although they were speaking English, I really never fully understood what they were saying. One conversation was particularly engaging for me as an outsider: this is where I learned about the concept of a “magic number.” It’s kind of a calibration number; it could be zero but may not be, it can show up out of the blue and may have no real meaning, it might be specific to only this robot and can be critical to know in order to calibrate the robot.
On another occasion I accused Marcel and Max of making up the term “dongle” to describe a piece of hardware on the back of the control unit. But it really is a word, probably arbitrary in its coinage (did someone think it was more descriptive than “thingy”?). As it turns out, it refers to a piece of hardware in the computer industry that acts like a key. Without the dongle the program/robot will not run; in this robot’s case, the dongle is attached to the auxiliary emergency-stop switch.
At the conclusion of the loan in Montreal, the robot will return to the AGO. I’ll need to establish a maintenance program and schedule for the robot and the installation’s various components; the arm should be occasionally manipulated, the compressor operated and the conveyor motor occasionally turned to prevent seals from desiccating and leaking. I’ve also requested that some space be found at the Gallery in order to install at least the robot and control unit so we can continue refinements of the program while the process is still fresh in our minds.
One of the most challenging aspects of the project was the identification and, in this case, numerical qualification of the movement in part and in whole — in other words the “performance” of the piece. The robot is capable of subtlety, and its actions should be precisely replicated each cycle, unlike a human performance, when there will be variations between each cycle. We spent many hours discussing and adjusting each joint, especially the wrist joints, where the movement could be most fluid and very delicate. In the end, however, the robot’s performance is still only as good as the information its human programmers can give it.
Curious about Conservation?
If you have a burning question about Conservation, leave a comment below. We’ll do our best to give you an answer in an upcoming Conservation Notes post.