ProbXiv
Explorersign in
Problem archiveProblem record

Statement

Hempel’s distance function was chosen so that it would capture the geometry, and indeed it does that very well, yet in some ways it feels unnatural. The Hatcher-Thurston complex HT(S) seems much more natural to us, since and pairs of vertices in the latter determine a Heegaard diagram, and one gets every genus g Heegaard diagram this way. One wonders whether it is possible to redefine Heegaard distance, using HT(S), or perhaps even P(S) or one of the other complexes that has proved to be so useful in studying subgroups of M, and whether new things will be learned that way? We have focussed our discussion, up to now, on the 3-manifold that is determined by a choice of an element φ in the group M via the Heegaard splitting construction. A very different construction which also starts with the choice of an element in the mapping class group, say α∈M g, produces the mapping torus of α, i.e. the surface bundle ( S× [0, 1])/α, defined by setting ( p, 0) = ( α(p), 1). Surface bundle structures on 3-manifolds, when they exist, are also not unique. Two surface bundles ( S×I)/α, (S×I)/α′ are equivalent if and only if α,α ′ are in the same conjugacy class inMg. In [ 1] an interesting description is given of a natural way to produce, for each ( S×I)/α, a related Heegaard splitting H∪βH ′. Choose a fiber S of (S×I)/α, say S×{ 0} and choose points p,q∈S, p̸=q, p̸=α(q). Let P andQ be disjoint closures of regular neighborhoods of p×[0, 1/2] and q× [1/2, 1] respectively. Set H = (S× ([0, 1/2]−Q)∪P, H ′ = (S× [1/2, 1]−P )∪Q. 152 J. Birman Note that H and H ′ are homeomorphic handlebodies of genus 2 g + 1 which are embedded in (S×I)/α and identified along their boundaries, so they give a Heegaard decomposition of ( S×I)/α. We call it the bundle-related Heegaard splitting of ( S×I)/α. It is H∪βH ′ for some β∈M 2g+1. We have several problems that relate to this construction:

Record

Added
Links

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.