Package for simplifying polynomials on non-commutative algebra

[awaw...@gmail.com]

Hi fellow Gophers

I wonder if anyone knows of a package that simplifies polynomials on non-commutative algebra?

As a concrete example, let `a` and `b` satisfy the commutator [a, b] = ab-ba = 1,

I want to simplify (a+b)^4 into aabb + ab + ...

There are libraries such as NCAlgebra and Bergman that do this using Gröbner basis.

I wonder does anyone know of something similar in Go?

Thanks!

[awaw...@gmail.com]

I have created this noncommutative algebraic geometry package to perform this task.

[Jason E. Aten]

Thanks Fumin. From looking at the test suite, this looks like useful and high quality work.

[awaw...@gmail.com]

Thanks Jason the kind words, I hope it is of some help to future like-minded scientists who enjoy Go.

I'd also like to share that I find it more productive to do science in a dull "systems language" like Go, compared to other more expressive, feature rich language like Python, Julia, or Mathematica. In fact, the increased productivity comes from better readability/maintainability, interoperability, and performance.

[Jason E. Aten]

I find Go very pleasant for things where the 

computation will run for much longer than it

takes to write the code.  Especially when you need

to leverage multiple cores in a non-uniform way, it

is hard to beat.

On the flip side, I find R offers 10-100x faster

 prototyping and exploratory data analysis (I mean the human time-on-task) 

when you need plotting and/or clustering; and Julia offers 

10-100x performance for some models that really benefit from 

inlining a computation kernel--Julia can even GPU-ize automatically.

I'm also happy to import Go code into R when needed;

https://github.com/glycerine/rmq

https://github.com/glycerine/rbook

https://github.com/glycerine/embedr

[awaw...@gmail.com]

Jason, those are nice observations. Also, pretty cool that you can import Go into R, learnt a new thing today. : )

[Dan Kortschak]

On Tue, 2025-12-23 at 19:37 -0800, Jason E. Aten wrote:
> https://github.com/glycerine/embedr

Also, https://github.com/rgonomic/rgo/

[awaw...@gmail.com]

Cool library, Dan, this looks like a principled and automated way of embedding Go into R.