0.30 As transparent as infrastructure: on the research of  (Page 11/13)

Reinvention is sustained in play not structure

Text analysis tools as an example

The boundary between technological and organizational means of information processing is mobile. It can be shifted in either direction, and technological mechanisms can only substitute for human and organizational ones when the latter are prepared to support the substitution. ( Understanding Infrastructure , p. 3)

I give, as an example of boundary moving, a genre of research tool I have been involved in: text analysis tools. I do this in order to demonstrate the fluidity of definition that can move such work from research to infrastructure and back again. I will do it by talking about selected projects that have attempted to develop text analysis tools.

PRORA output

PRORA. In 1966, the University of Toronto press published Glickman and Staalman’s Manual for the Printing of Literary Texts and Concordances by Computer. The manual covered the operation of PRORA, which was a mainframe batch concordance generation tool like OCP. PRORA was developed by a humanist (Glickman) and an engineer to facilitate the preparation of print concordances. The research tool was the print concordance; PRORA was a tool to support researchers like Glickman.

TACT screen

TACT. TACT, released in 1989 by the University of Toronto Centre for Computing in the Humanities, was one of a number of projects that set out to develop an accessible interactive text analysis tool in the 1980s. The tool was, and still is, free for download and the MLA published the manual, Using TACT. Using TACT can now, courtesy of the MLA, be downloaded as a PDF. The software that runs on MS DOS is available at (External Link) . TACT was designed to be usable on an IBM PC running MS-DOS and is still used with DOS emulators. The model was to build a widely useful and interactive tool that could run on what most humanists had on their desk. No one called tools infrastructure back then; it was just a research tool funded by a university centre. TACT was also one of the first tools that was designed to be used for interpreting on the PC interactively. You didn’t run a text through a batch process and then study the resulting print concordance, you used the tool in research. ARRAS (Archive Retrieval and Analysis System) was probably the first interactive concordancer. Smith writes about it in “A New Environment For Literary Analysis,” describing his model thus: “ARRAS should not be thought of as a ‘black box’ into which one inserts a text along with a set of commands and out of which one receives a completed analysis. A better analogy is a toolbox containing a set of tools, each designed for a particular task. The ARRAS design always presumes a human inquirer at the center. Thus ARRAS amplifies, rather than replaces, specific perceptual and cognitive functions” (p. 22). The tool instantiated changing ideas about how computer-assisted interpretation would take place. Finally, TACT was co-developed by a team with professional programmers and academics associated with the Centre. John Bradley and I adapted it to the web in the TACTweb project in 1998. See (External Link) .