Beowulf on Scantron

Beowulf on Scantron

Beowulf

Digitizing the Old English Corpus "In general, optically scanned systems are labor-efficient but introduce a proportion of machine errors, while human entry systems are labor-intensive, and vary in their accuracy depending on the people themsleves. In the early seventies [...] we rekeyed the entire corpus (that was almost the only means of data entry then) and our secretary, Elaine Quanz, spent most of six years typing it. But because time on the mainframe computer was very expensive, she didn't key the texts into an on-line file, but rather typed them in a special IBM Selectric font, OCR-B, which was legible to a primitive optical scanner. [...] The old-fashioned system did not function too badly, chiefly because of the extraordinary speed and accuracy--to say nothing of the good nature--of our secretary. The texts were typed one by one, coded with their alphanumeric text codes and entered with substitute characters for Old English æ, þ, ð, and ę. They were collected in batches and sent off to be scanned." (Ashley Crandell Amos, "Computers and Lexicography: The Dictionary of Old English," Editing, Publishing, and Computer Technology, AMS Press, Inc., 1988).

Photographer: Alexandra Bolintineanu