Yes, I'm aware of that. I still don't count this as a string theory success. Borcherds proof of the Moonshine conjecture uses the no-ghost theorem in a way wh

Yes, I'm aware of that. I still don't count this as a string theory success. Borcherds proof of the Moonshine conjecture uses the no-ghost theorem in a way which is completely unsurprising to anyone familiar with QFT. Perelman's proof of Geometrization, in contrast, uses the string-frame Lagrangian in a manner which is by no means implicit in the string theory literature. The deep content of those papers comes after he's decided which functional to work with.