My work with P. Ling and B. Kagstrom at the University of Umea in Sweden on GEMM-based BLAS moved to completion with the final production of our library. With our codes, manufacturers need only have a fast general matrix multiply kernel in order to implement all the structured matrix multiply kernels that make up the level-3 BLAS library. These codes make it possible to extract high performance when running LAPACK.
I continued to work on applications of the Kronecker product with Computer Science graduate student Nikos Pitsianis.
I produced a Matlab environment that supports the fitting of epicycles to planetary data, with Computer Science major Nikola Valerjev. We showed that if Copernicus had our system and was a poor scientist, then in all likelihood he would have preferred to patch up the epicycle model instead of going the heliocentric route. A video of our work is being marketed by SIAM, and high school students interested in computational science are the prime target.
The Pascal version of my Computer Science 100 text was published in Spring 1996. A Matlab version is currently being developed. The third edition of "Matrix Computations" (with G.H. Golub) is in production. Also, my CS 222 text, "Introduction to Scientific Computing: A Matrix-Vector Approach Using Matlab", is in production and should be available for fall adoption.
If you have questions or comments please contact: www@cs.cornell.edu.