Indeed, if Ringe, Warnow, & Taylor's computational model for the relationships between the Indo-European languages is correct, English is more closely related to Farsi than to French.
Computational Linguistics looks so cool. I started reading this and then I realized it was 71 pages, and I have way too many other things I'm supposed to be doing.
(That paper is technically "computational linguistics", I guess, in that it's a linguistics paper using computational methods; but it's not what's usually referred to by the term "computational linguistics". It's more like, computational cladistics applied to historical linguistics.)