Thursday, March 10, 2011

Computational Stylistics... bioengineering is nothing compared to this!

package inspired.flash.linguistics
    In my search for the perfect object news feed I learned something interesting about myself.
   I think I should have gone into linguistics.. specifically linguistics as it relates to computers.  I never would have known this part about myself in college, as nothing I was ever told or taught gave me any hint into this field, but now that I have found it, I wish there was a way I could be a part of it.

Today, I discovered something that isn't even on Wikipedia...   What is worse, is that this field, or at least the idea of this field has been around since 1964.   They call it COMPUTATIONAL STYLISTICS.

The idea here, is that you can take some written text, and convert it into a long string of characters.  This long string, can then be parsed and analyzed, patterns found and a literary finger print can be found.  So far, they have used this to analyze Shakespeare plays, and to discover which writers he influenced, and which writers influenced him. However, I was thinking that this could also be used to detect plagiarism.  Further more, if it is used in schools early enough, it can be further used to watch the development of writing in a person, and possibly even detect further patters from schools and regions etc.  The amount of self awareness that people can have of their writing would be amazing.

I wonder if these tools can also be used for code, and if source code can be linked to it's author. Could encrypted code also be linked to it's author? The implications for cyber security is also amazing...

I'll have to ask my friend about this...

For the first time in my life, I wish I had the temperament to go back to school...  But I'll be honest, I think I'd just be frustrated there... rather, I wish there was a way for me to meet someone who knows how to turn this into a business, and let me learn on the job.