The equations for the moduli space of \(n\) points on the line (Q999251)

The authors study the equations of the moduli spaces \(M_w\) of \(n\) ordered points on the projective line \({\mathbb{P}}^1\). For any \(n\) and any weight vector \(w = (w_1,..,w_n)\), \(w_i\) positive integers, they state the following questions, each of which subsumes the previous: (a) What are the ``good'' equations intersecting \(M_w\) set-theoretically? (b) What are the ``good'' equations intersecting \(M_w\) scheme-theoretically? (c) What are the ``good'' generators for the ideal of all equations for \(M_w\)? Theorem 1.1, the first main result, answers the question (a): The space \(M_w\) is cut out scheme-theoretically by the sign, Plücker, and simple binomial relations, with a unique exception of \(w = (1,1,1,1,1,1)\), defining the Segre cubic threefold. Theorem 1.3, the second main result, answers the question (c): For any weight vector \(w\), the ideal of relations in the coordinate ring \(R_w\) of \(M_w\) is generated by relations of degree at most four. If all the weights are even, then the ideal is generated in degree two. In the course of the study the authors prove in particular a ``weighed'' generalization of a classical result of \textit{A. B. Kempe} [Lond. M. S. Proc. XXV. 343--359 (1894; JFM 25.0235.01)]: The lowest degree invariants generate the coordinate ring \(R_w\), see Theorem 2.3. Although the authors note that ``their argument is almost entirely elementary'', the argument seems to be highly techical. At the end of the introduction are discussed some related problems. It is of worth to notice that the question about finding similar satisfactory results for the sets of \(n\) non-ordered points on the line still remains open, see Remark 1.4 and also Remark 1.5.