Description of the original site-specific installation
(or, ideas that culminated in Blind Reading)

The space

This is the “Distributed Collaborative Laboratory” at the IDRC, on Level 2 at 205 Richmond. From across the street (240 Richmond in particular), its three windows are clearly visible: the three leftmost windows on the third row, counting from the bottom.

For students in Inclusive Design (INCD)—a little known program at OCAD—, this is their classroom and meeting room. Outside of classes and office hours, however, this space is inaccessible to students, even though it would also be their studio space if they needed any.

Braille, bits, and blinds

Although window blinds are not meant to convey information, by manipulating them in various ways, information can be conveyed. This can be thought of as a “covert channel.”

One way to convey information is by distinguishing between the “open” and “closed” states of the blinds, which can be thought to represent a binary digit, or bit. By successively manipulating the blinds into perceptible states of open and closed (representing 0 and 1, for example), one can theoretically transmit any message that can be represented by bits.

Computer encodings are not the only codes that are binary based: Braille, for example, can also be thought of as a binary code, consisting normally of 6 bits (called dots) in a Braille cell. The dots are read from the top-left corner, the left three dots top-to-bottom (1, 2, 3) then the right three dots top-to-bottom (4, 5, 6), and the knowledge of this numbered sequence of dots (as opposed to just the visual pattern) is what allows one to write in Braille using a stylus and slate.

In practice, an endless stream of ones and zeros is incapable of being separated into the correct cells. “Stop bits,” as they are called, need to be present between each cell for cells to be made out. In a real computer encoding, there will be an extra layer of encoding so that stop bits (which carry meta-information) can be distinguished from the real bits that carry information. However, in the case of window blinds, we can introduce a third state of “half open” to serve as the stop bit. Thus our encoding is no longer truly binary, but a ternary system.

Of course, the manipulation of blinds is ultimately constrained by physics: we cannot pull the blinds up or down too quickly, so there is a lower bound to how long it takes to convey one bit of information. Assuming, for example, that we need 10 seconds to pull the blinds up or down and another 5 seconds for a person to register whether the blinds are open or closed, then if we are using Braille as the internal encoding, and if we are using one window to transmit, and one non-bit pattern between encodings of Braille cells, it will take approximately 15 minutes to transmit just a 10-letter word.