FLUXUS is a new chat interface that dynamically generates shapes above a user’s avatar that represent what they have typed into the chat box. The system is language agnostic to a certain extent, and can accept input from any language that uses the standard ASCII character set. Based on differences in character usage and word lengths, different languages have a subtly different “feel” from each other in terms of the resulting graphic effects above an avatar.

My Role: For this project I worked on the translation algorithm to go from typed input to visual effects, as well as assisted in programming the various avatars. Initially we wanted to base the effects on phonemes, so that words that sound similar will appear similar in the resulting graphics. However, the only real way to do this accurately would be with a massive dictionary lookup table, and the system would not be able to handle simple things like misspellings, abbreviations or emoticons that are common in all online chat environments. As a result we took the simpler route of directly translating groups of consonants into given shapes, while combinations of vowels within a given word determine all of that word’s shape’s colour. This had the added benefit of allowing us to accept input from other languages that used a similar alphabet, as the system was only looking at letter usage rates and not whole words.

For more information, visit the FLUXUS website.