Dr Minor would read a text not for its meaning but for its words. It wasa novel approach to the task – the equivalent of cutting up a book word byword, and then placing each in an alphabetical list which helped the editorsquickly find quotations. Just as Google today ‘reads’ text as a series of wordsor symbols that are searchable and discoverable, so with Dr Minor. A manualundertaking of this kind was laborious – he was basically working as acomputer would work – but it probably resulted in a higher percentage of hisquotations making it to the Dictionary page than those of other contributors.

Envisioning the next wave of emergent AI

I recall being told by a distinguishedanthropological linguist, in 1953, that he had no intention of working througha vast collection of materials that he had assembled because within a few yearsit would surely be possible to program a computer to construct a grammar froma large corpus of data by the use of techniques that were already fairly wellformalized.

At the back of Dr Duncan's book on the topic, Index, A History Of The, he includes not one but two indexes, in order to make a point.