Just to be crystal clear, I need to say this. If I encounter somebody else telling me that "email is less than face-to-face and therefore it limits the ability of communicating" (or something to that extent), I yell.
Emails have less cues than face-to-face stuff. Emails have written words and some paralanguage to convey the message, for sure less than face-to-face. Does it mean face-to-face is a way of communicating that is less ambiguous?
I'm sorry for the mathematically-challenged persons, but more cues mean adding complexity to the mix, NOT lowering complexity. The higher the number of variables in an equation, the higher the number of solutions. DUH.
And I'm also sorry for the statistically-challenged persons. Supposing to factorize all the variables in the simplest possible way (true/false -- for sake of simplicity) and supposing the probability attached to those variables to be independent (like in the case of two persons not knowing each other and meeting -- aka the case of each person not knowing what a particular voicetone or smile means to the other person), the probability to accurately predict the meaning of a message with lower cues is actually higher than the probability to be able to predict the meaning of a message with higher cues.
1 cue: 50%
2 cues: 25%
3 cues: 12.5%
and so on.
Hence, the likelihood one has to correctly categorize/interpret the meaning of a message from an unknown person is actually higher through email than through face to face encounter.
The fact one usually thinks it would be the reverse is the same reason for people buy lottery tickets: overestimation of actual probability due to a cognitive bias.