(The revisions are identical or unavailable.)

A number of people have begun to discuss the implementation of a boost library for Clifford Algebra. This is a formulation of algebra which has advantages in some areas of calculation performance, but also provides a unified formulation of complex numbers, quaternions and vectors.

What has emerged for me at least is the fact that we have more than one good possible starting point, and that the best way ahead is

- To define the different purposes for such a library, as the emphasis for different users may be different.
- I am interested in a library which can handle Clifford objects of many different basis signatures, and also perform symbolic algebra.

- To look at what is available at this point and to take stock of what each author has done so far and is planning to do anyway.
- Then to look at some stages in which the code could be developed into a library which would satisfy the criteria for boost.
- I intend to look at the feasiblity of combining GluCat with Jaap's code. I am waiting until the release of Boost 1_29_0 with MPL to make a start on this.

Although I polled the Boost community for interest in a Geometric Algebra library, my current efforts (http://www.jaapsuter.com) should not be taken as if I wanted to make a definite proposal. I would really like to see a discussion where we talk about the trade-offs of different implementations. I think both nKlein and GluCat are amazing libraries that can certainly live up to Boost expectations, given they would be Boostified.

[Jaap Suter]?

Disclaimer: This site not officially maintained by Boost Developers