We would like to congratulate Israel Moiseevich Gelfand on the occasion of his 90th birthday and to thank the organizers P. Etingof, V. Retakh and I. Singer for the opportunity to take part in the celebration. The present paper is closely based on the lecture given by the second author at The Unity of Mathematics Symposium and is based on our joint work in progress on Gromov-Witten invariants with values in complex cobordisms. We will mostly consider here only the simplest example, elucidating one of the key aspects of the theory. We refer the reader to [9] for a more comprehensive survey of the subject and to [5] for all further details. Consider M 0,n , n ≥ 3, the Deligne-Mumford compactification of the moduli space of configurations of n distinct ordered points on the Riemann sphere CP 1 . Obviously, M 0,3 = pt, M 0,4 = CP 1 , while M 0,5 is known to be isomorphic to CP 2 blown up at 4 points. In general, M 0,n is a compact complex manifold of dimension n − 3, and it makes sense to ask what is the complex cobordism class of this manifold. The Thom ring of complex cobordisms, after tensoring with Q, is known to be isomorphic to U * = Q[CP 1 , CP 2 , ...], the polynomial algebra with generators CP k of degree −2k. Thus our question is to express M 0,n , modulo the relation of complex cobordism, as a polynomial in complex projective spaces.This problem can be generalized in the following three directions. Firstly, one can develop intersection theory for complex cobordism classes from the complex cobordism ring U * (M 0,n ). Such intersection numbers take values in the coefficient algebra U * = U * (pt) of complex cobordism theory.Secondly, one can consider the Deligne-Mumford moduli spaces M g,n of stable n-pointed genus-g complex curves. They are known to be compact complex orbifolds, and for an orbifold, one can mimic (as