Home

Welcome to the Lingualizer wiki!

Lingualizer is a set of tools for representing poems as highly detailed and coherent linguistic data-sets.

##". . . studying the incredible intricacies of the verbal art of the greatest masters of language without having the most subtle of linguistic tools to do one’s literary criticism with – this is as successful as doing astronomy with an inexpensive telescope." – Haj Ross

At the moment, Lingualizer is an odd assortment of command-line perl scripts, php, javascript, hand-rolled xml, mysql and plain text databases inspired by the empirical approach to Poetics professed by Haj Ross (http://en.wikipedia.org/wiki/Haj_Ross) in his course Linguistics and Literature. Despite it's being embarrassingly disorganized, it does have some functionality, including:

• a 100k+ word pronunciation dictionary with lexical stress (English only for now, but polyglot by design)
• automatic (English only) and interactive tokenization by stanza, line, word, orthographic syllable (lookup), phonologic syllable (computed), onset/nucleus/coda, and phonological segment
• basic descriptive metrics including: element counts, repeated lines, repeated words, etc.
• dynamic local and global sequence alignment of segment strings
• dynamic similarity/distance comparison of distinctive feature matrices, both for individual segments and for mixed length strings

This current project is a complete, from scratch rebuild of Lingualizer as a unified suite of Django data models and web-based tools.

Progress Note: None yet. First steps will be to recreate the existing mysql database in Django and populate the dictionary.