1: \begin{abstract}
2: We introduce a combinatorial property for
3: finitely generated groups
4: called \stkbl\ that
5: implies the existence of
6: %provides?
7: an inductive procedure for constructing
8: van Kampen diagrams with respect to a
9: canonical finite presentation.
10: We also define \astkbl\ groups, for which
11: this procedure is an effective algorithm.
12: This property gives a common
13: model for algorithms arising from both rewriting
14: systems and almost convexity for groups.
15:
16: We also introduce a new pair of asymptotic invariants that are
17: filling inequalities refining the
18: notions of intrinsic and extrinsic diameter inequalities for
19: finitely presented groups. These \tfs\
20: are quasi-isometry invariants, up to
21: Lipschitz equivalence of functions (and, in the case of
22: the intrinsic \tfi, up to choice of a sufficiently large
23: set of defining relators).
24: We show that the radial
25: tameness functions of~\cite{hmeiermeastame} are
26: equivalent to the extrinsic \tfi\ condition,
27: and so intrinsic \tfs\ can be viewed as the
28: intrinsic analog of radial tameness functions.
29:
30: We discuss both intrinsic and extrinsic
31: \tfs\ for many examples of \stkbl\ groups,
32: including groups with a finite complete
33: rewriting system,
34: Thompson's group $F$, Baumslag-Solitar groups
35: and their iterates, and almost convex groups.
36: We show that the fundamental group
37: of any closed 3-manifold with a uniform
38: geometry is algorithmically stackable
39: using a regular language of normal forms.
40: %We also bound \tfs\ for
41: %groups admitting normal forms with a
42: %fellow traveler property.
43: %For a group with a finite complete rewriting system,
44: %the string growth complexity function is (up to
45: %Lipschitz equivalence) an upper bound for both
46: %of these new invariants, and this result has
47: %application to the process of
48: %searching for finite complete
49: %rewriting systems for groups.
50: \end{abstract}