On the product theorem (Q1826303)

Faltings' product Theorem is a generalization of Roth's Lemma [\textit{G. Faltings}, Ann. Math. (2) 133, No. 3, 549--576 (1991; Zbl 0734.14007)]. It states that if the zeros of index \(\sigma\) of a multihomogeneous polynomial \(P\) have a component \(Z\) in common with the set of zeros of index \(\sigma+\epsilon\), then \(Z\) is a product. The most useful corollary is for \(\sigma=0\). After the work of \textit{M. Nakamaye} [Bull. Soc. Math. Fr. 123, No. 2, 155--188 (1995; Zbl 0841.11037)], \textit{P. Philippon} [Bull. Aust. Math. Soc. 59, No. 2, 323--334 (1999; Zbl 0927.11040)] has shown how to deduce this corollary from his zero estimate [Rocky Mt. J. Math. 26, No. 3, 1069--1088 (1996; Zbl 0893.11027)]. Effective versions due to \textit{J.-H. Evertse} [Acta Arith. 73, No. 3, 215--248 (1995; Zbl 0857.11034)] and \textit{R. Ferretti} [Forum Math. 8, No. 4, 401--427 (1996; Zbl 0860.11038)] give upper bounds for the degree and height of \(Z\). Here the author proves a new sharp effective version of this result by relaxing the hypotheses on the degrees of \(P\). The improvement compared with the previous estimates is due to the use of Samuel's multiplicity, like in Philippon's work, in place of the length, while the improvement with respect to Philippon's result is due to the use of a multiprojective height [chap.~5 and 7 of \textit{Yu. V. Nesterenko} and \textit{P. Philippon} (eds.), Introduction to algebraic independence theory (2001; Zbl 0966.11032)], instead of a reduction to the projective situation.