Methodology

In class we were presented with some new tools with which to help do research. However, some were difficult to use for my research (esp. TACT), but I have worked with others that have been more helpful with my research and data. The signif shell script tool (short for significance) was written using the unix grep, sed and awk tools. I have been beta-testing the tool using my data.

TACT was hard to use because it did not work well with large volumes of data. Pat and Xtract were both confusing and harder to use conceptually. TACT had the easiest user interface of the three.

The Net provided many tools to assist me in my research. However, by far the most useful were those tools which assisted in my communications with other people. Usenet News in particular helped me identify a pertinent and important topic to choose. I started the conversation off by posting ideas which developed for my midterm assignment in the class, and other people helped me identify that the question of who would be able and should be able to access of the developing network is of crucial importance.

In addition, e-mail was helpful in the transport of information which was highly important to my work. The Charles Babbage Institute is located at the University of Minnesota in Minneapolis. Normally I would not have been able to travel to Minnesota to utilize their archives for several reasons. The Interlibrary Loan System is often helpful, but has its own problems and complications. I was able to learn which interviews they had via email, and eventually get copies. However, it did take a while for the Institute to send me e-mail copies of the interviews. It might have been one of the first times they have sent out copies via e-mail. E-Mail also provided me the capability to easier handle the Interviews I acquired from CBI. The interviews were in machine readable form, so I could easily search the text, make a data base of them, and so on.

LINK to response from NIH.

Lastly, I utilized some unix tools to help organize and search through the madryid of information the Interviews represented. I was able to pinpoint the sections of the interviews that were of most importance to my presentation via using grep, sed and awk. The tool is called signif and upon supplying it with regular expression search terms, it is able to give an index of locations in various files, and then to provide a 10 (or user-defined) line contextual surroundings. Signif is still in the developmental stages.

The unix shell Script I used as a tool. The man page for this tool.

Here is some sample output from the tool signif.