To expand: the paradigm of only offering +1 and not -1 is, to me, akin to the whole "everybody's a winner" problem; if I go to a product page, and all it has is a +1 button, and the 'score' is 0, then there's absolutely no context to justify the score.
Is it that the product is brand-new (and/or just added the +1 button), or conversely, is it that the product is horrible and though the +1 button has been there a year, nobody has clicked it?
Whereas if I were to visit and I could see that the score was -1000, then that would give me a very good indication that the product (or some aspect of it) sucks.
Then of course there'd likely be situations where there'd be a mix, and so it'd be good to see a representation that took all votes into account, e.g. [+1000/-500]; with that, I could see that the sample size is moderately sufficient (1500), and that most people approve -- but that it's not without detractors.
Mostly, though, I'm tired of "everybody's a winner".