Hey, nice though experiment! Would be an exciting use of all the big data being produced.
Mohit Patni

Great question! 
(a) Subjectivity: The “scores” on any of these dimensions is pretty complex because each of of these traits have sub-traits. For example, extraversion has sub-traits such as activity, sociability, shyness and so on. Even if we disregard the correlation between traits and sub-traits, it is still difficult to know impact of trait/sub-trait to the whole personality. In short, yes, the scores would be heavily subjective.

(b) Homogeneous personalities: I do not have any data to prove or disprove. But, since the genetic makeup can’t be changed in a human, a majority of the personality can’t be changed easily. Also, we can only optimise to a certain extent. If a person is low on say, agreeableness, we could make little more agreeable than what he/she is. Making them a yes-man or yes-woman would still be close to impossible.

But then, my guess is only as good as yours.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.