Ok, again. Taxicabs are not exciting, not scaleable, and involve a lot of machines and humans to make them work.

How many iterations do you have in mind for this?

EDIT: My answer for "can a rapidly growing start-up successfully take on taxicabs?" is "yes."