A review of the paper

Summary

In this paper, the authors extend the finite length property from the two previously known cases of the countable pure set and the countable dense linear order without endpoints to two broader classes of structures. An infinite structure has the finite length property over a given field if, for each of its finite powers, chains of equivariant subspaces in the corresponding free vector space are bounded in length. The authors develop two methods for proving this property.

In Theorem 5.4, they prove the finite length property for those structures A – e.g. the equality atoms and the bit-vector atoms – which the authors call oligomorphic approximation, is a relaxed version of smooth approximation known from model theory. This is under an additional assumption that the underlying field has characteristic 0.

In Theorem 7.3, they prove the finite length property for those structures A – e.g. the ordered atoms – which arise as the generically ordered expansions of the Fraïssé limits of free amalgamation classes, over finite relational vocabularies of arity at most 2. Here they do not restrict the underlying field.

The Rado graph and related structures can consequently be treated using both approaches. The paper also discusses the motivation for orbit-finitely spanned vector spaces and their connections with weighted orbit-finite automata, function spaces, and systems of linear equations. The authors conclude by discussing several open questions, including whether every oligomorphic structure has the ascending chain property and whether every homogeneous structure over a finite relational vocabulary has the finite length property.

Opinion

The main strength of the paper is that it provides two different techniques for establishing the finite length property and applies them to classes substantially broader than the examples previously known. The overlap between the two methods, in particular for the Rado graph, is also useful because the methods have different assumptions, especially concerning the characteristic of the underlying field. The overall organisation of the paper is clear, and the outline given in the introduction helps the reader understand the structure of the argument. Figure 2 was particularly helpful, as it provides a proof sketch connecting several of the intermediate lemmas and theorems.

On the other hand, some aspects of the exposition could be made more accessible. The definition of an orbit-finite structure/vector space was not completely clear to me, particularly the role of quotienting by an equivariant equivalence relation. A concrete example illustrating this construction would make the definition easier to understand. More generally, some of the definitions are introduced quite abstractly, and additional examples could help motivate them. For example, the discussion of weighted orbit-finite automata could benefit from a small concrete example showing how the abstract vector-space framework arises in an automaton.

Questions

Since the Rado atoms satisfy the hypotheses of both Theorem 5.4 and Theorem 7.3, what additional insight is gained from having two proofs of the same result? Beyond the different characteristic assumptions, do the two methods suggest different generalisations or applications?

The paper introduces automorphisms and then explains orbits through the action of the automorphism group on tuples, with a graph example. However, since group actions and orbits are used extensively throughout the paper, would a brief formal definition of these notions before Definition 2.1 make the exposition more accessible to readers less familiar with group actions?

Is there a general criterion, preferably one that is effectively checkable, for determining whether a given oligomorphic structure admits an oligomorphic approximation? Theorem 5.2 establishes this property for several important examples, but it is not clear whether there is a broader characterization of structures admitting such an approximation.

Theorem 5.4 requires characteristic zero, whereas Theorem 7.3 applies without restricting the characteristic of the underlying field. Where is the characteristic-zero assumption essential in the first proof, and is there a known obstruction to extending this method to positive characteristic?