INTRODUCTIONOur goal in this paper is to identify certain naturally occurring colimits of schemes and algebraic spaces. The statements, which are essentially algebraization results for maps between schemes and algebraic spaces, are elementary and explicit. However, our techniques are indirect: we use (and prove) some new Tannaka duality theorems for maps of algebraic spaces. Our approach to these theorems relies on a systematic deployment of perfect complexes (ergo, we use some derived algebraic geometry) instead of ample line bundles or vector bundles. Consequently, the Tannaka duality results we obtain have fewer, and much weaker, finiteness constraints than some of the existing ones: we only insist that our algebraic spaces be quasi-compact and quasi-separated (qcqs), and do not require any quasi-projectivity or noetherian hypotheses. All rings are assumed to be commutative.1.1. Algebraization of jets. The first colimit we identify is that of an affine (adic) formal scheme.Theorem 1.1. If A is a ring which is I-adically complete for some ideal I, and X is a qcqs algebraic space, then X(A) ≃ lim X(A/I n ) via the natural map.An equivalent formulation is: if A = lim A/I n , then Spec(A) is a colimit of the diagram {Spec(A/I n )} in the category of qcqs algebraic spaces. Theorem 1.1 is straightforward to prove if A/I is local; its content becomes apparent only when Spec(A/I) has some non-trivial global geometry. Note also that there are no noetherian assumptions on any object in sight, so the ideal I might not be finitely generated. In fact, the result extends to more general topological rings A that arise naturally in p-adic geometry (see Remark 4.3). This answers a question asked by Drinfeld, and has the following representability consequence in the theory of arc spaces, which was our original motivation for pursuing Theorem 1.1. Corollary 1.2. If X is a qcqs algebraic space, then the "formal arc" functor Arcs X (R) := X(R t ) is an fpqc sheaf on the category of rings, and is identified with the functor R → lim X(R[t]/(t n )).As the functor Arcs X is almost never locally finitely presented (even for X an algebraic variety), one cannot reduce Corollary 1.2 to the corresponding assertion on the category of noetherian rings (which is easier to prove). This corollary answers a question raised in [NS10, §2] and pointed out to us by Nicaise. The following feature of the proof of Theorem 1.1 seems noteworthy: given a compatible system {ǫ n : Spec(A/I n ) → X} ∈ lim X(A/I n ), we construct an algebraization ǫ : Spec(A) → X without ever musing about points of Spec(A) \ Spec(A/I).
Algebraization of products.The second result deals with products, rather than cofiltered inverse limits, of rings; this question was brought to our attention by Poonen.An equivalent formulation is: the scheme Spec( i A i ) is a coproduct of {Spec(A i )} in the category of qcqs algebraic spaces. Note that some finiteness hypothesis on X is necessary: the (typically non-quasicompact) scheme ⊔ i Spec(A i ) is a coproduct of {Spec(A i )} in the cat...