sagemath/sage

Add tensor_factors() method to TensorProductOfHighestWeightModules in quantum_group_gap

Closed this issue · 13 comments

This fails with gap_packages installed since #34393.

However, there is another failure for the submodules. I do not think subobjects of tensor products should have a tensor product structure.

CC: @mkoeppe @fchapoton

Component: algebra

Keywords: quantum group, representation

Author: Travis Scrimshaw

Branch/Commit: 918c372

Reviewer: Matthias Koeppe

Issue created by migration from https://trac.sagemath.org/ticket/34635

Replying to [ticket:34635 Travis Scrimshaw]:

I do not think subobjects of tensor products should have a tensor product structure.

I disagree

But it should be documented

comment:2

If I take V (x) W and the submodule V (x) w for some fixed w \in W, what is the tensor product structure on that? I don't see how to make sense of that.

comment:3

What are the morphisms in the category of tensor products of modules?

comment:4

There is no difference with modules. Are you trying to say that we should have the tensor factors of V (x) w be V and the span of w? That might work, but how would this work with the problem at hand, where we have an A-repr inside of V (x) W?

comment:5

Replying to Travis Scrimshaw:

There is no difference with modules.

Let me elaborate slightly on this. I believe we agree that the tensor product of finite dimensional modules is equivalent to the category of modules (over a (commutative?) ring R). It just has a little extra information about how the module was constructed (which is important for things like graded modules and representations).

comment:6

Anyhow, an optional package gap_packages ought to pass the doctests, and it does not at the moment. Can you at least silence these warnings?

Commit: 918c372

Author: Travis Scrimshaw

comment:7

Here is a version where I work around the submodules of tensor products thinking it is a tensor product: I weakened the category of the submodules.


New commits:

67340e2Adding tensor_factors() to tensor product of quantum group modules.
918c372Changing the category of the submodule to get around subobjects of tensor products being in tensor products.
comment:8

LGTM. Tests OK with gap_packages installed.

Reviewer: Matthias Koeppe