Classics revisited: Éléments de géométrie algébrique (Q1621509)

The author provides a mathematical description of context and content of Éléments de géométrie algébrique (ÉGA) by \textit{A. Grothendieck} [Publ. Math., Inst. Hautes Étud. Sci. 4, 1--228 (1960); ibid. 8, 1--222 (1961); ibid. 11, 349--511 (1962; Zbl 0118.36206); ibid. 17, 137--223 (1963; Zbl 0122.16102); ibid. 20, 101--355 (1964; Zbl 0136.15901); ibid. 24, 1--231 (1965; Zbl 0135.39701); ibid. 28, 1--255 (1966; Zbl 0144.19904); ibid. 32, 1--361 (1967; Zbl 0153.22301)], whose first chapter has got the new edition [Éléments de géométrie algébrique~I. Berlin: Springer-Verlag (1971; Zbl 0203.23301)]. He focuses on the notion of scheme as ``geometric object'' and on two celebrated applications: the Weil conjectures solved by \textit{P. Deligne} [Publ. Math., Inst. Hautes Étud. Sci. 43, 273--307 (1973; Zbl 0287.14001); ibid. 52, 137--252 (1980; Zbl 0456.14014)] and the Birch and Swinnerton-Dyer conjecture. He presents schemes as an answer to the shortcomings of algebraic varieties at describing the solution set of a system of polynomial equations. Instead of restricting to rings of functions, the ÉGA propose to associate a geometric object, the ``affine'' scheme, to any ring, and to glue these objects together into general schemes. The author describes how category theory gives the right hints for defining operations and properties for schemes, e.g.\ flatness, smoothness, fiber product, intersection, finite-type morphisms. He explains how a classification problem of schemes may also be solved as a geometric object, the ``parameter space'', also called ``moduli space'', together with morphisms of the schemes into it; Yoneda's lemma shows that it is enough to devise a functor from the category of all schemes to the category of sets and to show that it is representable. The concept of scheme turns out to be too narrow for this to work in general, and the author addresses three strategies in this respect: ``algebraic stacks''; ``coarse'' moduli spaces; ``rigidification'' which consists in modifying the problem by fixing additional parameters (see Definition~5.1 and Theorem~5.2).