~~~
I hover over a scanner, turning physical page to digital particulate. I marvel at how images transmute into text, made searchable by machine. Optical Character Recognition is a built-in function that occurs with the scanning of a document, the process wherein images of text, handwritten or otherwise, are identified and converted into machine-encoded text. Jagged pixellate edges coalesce into legible unicode forms, meaning sublimated from black etchings, saturating the air with a dark, pungent smell.
As with many systems, it is not perfect. Recognition does not occur flawlessly, and errors emerge with impish regularity, a mischievous child scrawling with crayon on the wall behind the sofa. During Optical Character Recognition, the characters occasionally blur, melding into each other or fading into the grain of paper, numbers becoming letters, letters becoming other letters, and then becoming nonsense punctuation. The text becomes an abstracted version of its former self, an endearing attempt at coherence. Ins1ead ,’,’e f:nd oursclvcs lcf- vv:th t000 ,’,’ays :n vvh:ch a scntcncc ma) losc :ts or:g:nal-mcan:ng}
Optical Character Recognition is the attempt to identify potential markers of meaning, hypostasising these signs into coherent and accessible systems of representation. I softly apply pressure as documents sensuously smoothen against the glass flatbed of the scanner, bodies of knowledge spread across a lustrous surface, anticipating permeation by shards of light. I feel like I may be falling in love with a ghost in a machine.