A typical language survey may involve activities like determining linguistic relationships through the comparison of word lists, testing dialect intelligibility by playing back tape-recorded texts, and studying sociolinguistic aspects of language use and language attitudes in multilingual situations. WORDSURV (for Word Survey) is a computer program designed to aid in the first of these areasthe collection and analysis of word lists.
WordSurv Version 2.5, (MS-DOS) was developed by John S. Wimbish in the 1980s. It was first tested with Zambales data from the Philippines (Wimbish 1986), which involved 50 word lists that were 372 items in length. It is written in the C programming language and is compiled for IBM PC-compatible computers (which use the MS-DOS operating system).
The phonostatistic method used in WordSurv 2.5 is based on a quantity called the degrees of difference. The COMPASS algorithm (Comparativist’s Assistant), developed by Donald Frantz (1970), is used to determine the genetic relationship between languages (Comparative Method).
Transcriptions are stored using combinations of Keyman characters that were displayed as single phonetic characters once the correct font was installed. It supports special characters only by supporting the standard extended character set available on IBM-PC compatible computers. This extended set includes the 94 printable characters of the regular ASCII (American Standard Code for Information Interchange) character set, plus 128 more printable characters, which include (among other things) accented vowels, Greek letters, and mathematical symbols. With certain kinds of video adapters, such as the EGA and the Hercules Plus, it is possible to download user-defined shapes for the characters. WordSurv 2.5 will support such user-defined character sets. Even if you are not able to get the desired character shapes on the screen, there are many word processors which can take a file that uses the extended character set (such as a WordSurv output file) and replace the program's characters with user-defined ones at print time. In this way special characters such as those used in phonetics may be displayed on the printed page, even though they cannot be displayed on the screen.