Huh? That is to say, I work on the Midrange Storage Management software development team at CLARiiON. CLARiiON used to be a part of Data General. Data General was bought by EMC Corp., so now Data General and CLARiiON are sub-companies of EMC Corp. Get it?
This was an Undergraduate Senior Project, and of a scope appropriate to that. It was not, in fact, a Master's Thesis or equivalent, and I did not have an expert knowledge of the subject mater or previous experience with it. In an ideal world I could have been an academic and a researcher for life, but if that were my career, it would be in the field of Psychology, not Computer Science. I don't intend to pick this up again, or do any further research in the area, though part of me would really like to. There are some tests in particular that I would like to redo in greater depth, as well as experiments left untried, etc. My advisor consoled me at the end of my project by telling me the secret of research - nothing is ever finished. And if you are a perfectionist, nothing you do will ever satisfy you completely, and you will probably be suspiciou of people who tell you you have done great work.
There was a specific question in the email query - something like whether or not I had considered using some property of Inverse Fourier Transforms. Something like that. If you think it would apply in an interesting way, run with it, but I have to say that no, I didn't. My knowledge of Fourier Transforms and how to apply them to different problems is totally limited to this simple undergraduate project and the research I did for it (in which I never encountered mention of using the inverse property of Fourier Transforms). I have no idea what effect that would have on the transforms or the matching, but it sounds like fun.
The researcher from California suggested that I might have been the first person to think of using a Walsh transform on categorical data in the way that we did (transforming multiple sequences of binary data representing each of the symbols of the original categorical sequences, etc). I don't think so, for two reasons. First is that my advisor, Professor Alvarez, thought of it in the car on the way to one of our meetings (during which the three of us fleshed it out further), so if anyone thought of it first it was him. Second, I think someone thought of it before we did, though we didn't know it at the time. We encountered an article in which someone was describing the very same idea for use with the same kind of data sequences (DNA)! I was excited and disappointed simultaneously when I saw that article. I'm sorry that I don't remember the author or the publication.