A Guide to Using the Minimal Coding Programs

Clifton Pye

The programs that perform the analyses outlined in the document Minimal Coding can be downloaded from this site. They are dos programs, which means that they must be run from the dos window on a pc running Windows 95/98/me. I have run into some difficulties using the programs on pcs running Windows nt so users are advised to check that their computer can run the programs. Sorry, a version of the programs for Macintosh and Unix is not available.

Before starting the programs, make sure that the analysis program is in the same folder as the data file you wish to analyze. You should be able to run the programs by double clicking on their icon. The programs usually request two pieces of information: 1. the name of the data file, and 2. the code of the speaker you wish to analyze. You need to be careful about both of these entries. The programs disregard case so you can use upper or lower case. As dos-based programs, however, the name of the data file is restricted to 8 or fewer letters plus a 3-letter extension, e.g. “filename.txt” Be sure you type in the name of the file correctly, including the 3-letter extension if there is one. Any deviation from this naming convention (e.g. spaces or more than 8 characters) will cause problems and force to you close the program window while the program is still running. The programs should only take half a second to process the data files so it is a good idea to close a program window after a second or two.

The second part of running the program is to type in the subject code correctly. I have designed the programs to work with subject codes that contain or do not contain colons (‘:’), tabs, or spaces in addition to the subject codes. Subject codes can be any of the following:

abe my phrase.

a: my phrase.

a my phrase.

ab: my phrase.

Since I am an advocate of minimal coding, a single letter and space should be sufficient. It is not necessary to type the colon, space or tab when the programs ask for the subject code. Simply type in the letter(s).

The programs are Addtier, Palabra, Verbs and Qancord. Addtier adds a morphological tier after every subject line. I assume that all subject lines are contained on a single file line, and do not continue on a following line. You can start Addtier, by double clicking on the Addtier icon. The program will open a dos window and prompt you for the name of your data file and the code name for the subject’s utterances in the file. The program will add an ‘equivalence’ line to any subject line that does not already contain an equivalence line. I use equivalence lines to show the difference between child and adult productions, e.g.,

chi two bus.

= two busses.

Each equivalence line should begin with a space, an equal sign, and a space. Addtier preserves equivalence lines that exist in the original data file. The program will produce its output in a file with the same name as your input file plus the file extension ‘.out’. If your input file was named ‘data.txt’ Addtier will produce an output file with the name ‘data.out’.

Palabra produces a lexical concordance from the files created by Addtier. Every subject line must be followed by an equivalence line for Palabra to run correctly. You should be able to run Palabra directly on the file produced by Addtier without coding for morpheme boundaries. Palabra will produce an output file with the name ‘data.wrd’.

Qancord will produce a morphological concordance for files with morpheme boundaries marked in the morphological tier coded ‘%mor’. Qancord uses a space or the delimiters -.,;?!/)<>'#+\* as morpheme boundaries and/or morpheme error markers. In other words, it will count any sequence between a space and a hyphen (‘-’), or between a comma (‘,’) and a space, etc. as a morpheme. The program assumes that any characters that follow a parenthesis (‘(’), a square bracket (‘[’) or a curly bracket (‘{’) are comments and will exclude them from the concordance. For this reason, I advise using angle brackets (‘< >’) to enclose parts of a speaker’s utterance that will be included in the output of the concordance. Qancord uses the same file name conventions as Addtier, i.e., Qancord assumes that the data file will only have a file name and an extension. If you wish to use the Qancord program on the file that Addtier produced, you must include the extension ‘out’ for this file. Qancord also assumes that every line for the subject contains an equivalence line so the best practice is to run Qancord on the output from Addtier. Qancord produces an output file with the extension ‘.con’. If you input a file named ‘datafile.out’, Qancord will write its output to the file ‘datafile.con’.

I include an additional program named ‘verbs’. I have found a program that extracts utterances containing verbs to be extremely useful. The program ‘Verbs’ assumes that you have marked the beginning of all verb roots with a forward slash (‘/’). The program then produces a Qancordance file for all of the utterances that contain verbs indicated by the forward slash. This program once again assumes that the input file name does not have an extension. It produces an output file with the extension ‘.vbs’.

These are ‘bare bones’ programs that follow the minimal coding paradigm. Additional bells and whistles could be added, but at the cost of adding additional complications to the coding conventions. I would always appreciate any comments or suggestions for improving the programs.