Call and Response - Accidents
How do you manage expectations when working with a machine? I keep wondering how to plan for unexpected deviations, how to draw with no assumptions, and how to reflect on how I have been relying on the act of translation across machine to erase my assumptions about what it draws.
A moment of lead breaking and being lowered mid-drawing
It seems that this translation isn’t big enough to take care of this, to produce something unexpected. No, I know where the line will be drawn, what color it will be (since I am the one who loaded the pencil), and I have learned the machine’s quirks enough to expect certain deviations.
We define accidents as unexpected - and at first glance, it seems like drawing with this instrument is full of them. When it only moves in one direction instead of diagonally along two axes at once, the lines pair up together, every other one spaced differently than how it is drawn in the computer, and the instrument is unaware of this miscalculation of space. Or, the lead breaks when the pencil is lowered along the z-axis, or it lightens as the tip is dulled over the course of a drawing - the instrument doesn’t realize these things either.
But when does familiarity with standard machine “accidents” make them non-accidental, or expected? And if you can anticipate an accident, what do you call this known deviation?
Outside of giving instructions, I, as the operator, have a few more moments of influence: how far the lead is clicked out before the instrument starts to draw, where I tell it the origin of the x, y, and z axes is. If the x and y origins are set too far off, the machine will motor outside its limits and try to escape its aluminum frame/world. If the z-origin is set too high, the pencil will never touch paper, and if too low, the lead will break.
Precise instructions do not guarantee absolute certainty (McVicar, 240), but the uncertainty in this collaboration comes from human errors, not machine. Randomness occurs outside of this instrument, in the creation of the code that turns lines into numbers and my own manual adjustments of axes, lead, and paper. But even “randomness” here is not random at all, but rather optimized to minimize machine movement, through an open source laser cutting software that I tricked into sending the same message for turning a laser on and lowering a pencil, and filtered through my own intuition.
This machine, then, maybe cannot act with randomness, uncertainty, or accidents. It follows instructions in the same blind way, unknowing of output or consequence, regardless of where it is in space or what tool it is holding. It deviates in known ways, producing the same accidents over and over until they can no longer be considered accidental, becoming a kind of cadence or handwriting to how the machine draws.
Paired lines, the Tiny-Z’s most consistent “accident”
References:
McVicar, Mhairi. 2019. Precision in architecture: certainty, ambiguity and deviation.