black project
In the analysis of a natural language's phonology (the system of sounds occurring in this language and their interactions), a prevalent approach is to view each sound (or segment) in the language as a "feature matrix". This structure defines a series of phonological features that describe the state and movement of the vocal tract during the production of a segment. Based on whether a segment exhibits the particular state and movement identified by a given feature, the segment is said to be positively or negatively valued for that feature. The transcription of a segment in a phonetic alphabet such as IPA thus functions as an abbreviation for the segment's complete feature matrix.
With such an approach, the change from one sound to another
through the course of history or between a plausible mental
representation and an articulatory realization can be described by
changing the former's feature values into those of the latter.
Given that the domain of each feature's possible values is binary,
this operation is not unlike setting and clearing bits in a bitmap
in computing.
Phonomask aims precisely to explore this parallel between disciplines
by investigating the effectiveness of representing feature matrices
with bitmaps and verifying phonological processes through bitwise
manipulation.