Credits
 
 

Aldor is the result of the work of many people. Here we outline the major contributors, both individuals and organizations. While we have attempted to acknowledge all major contributors, we may have made some inadvertent omissions. If you are aware of any, please directed inform us at credits@aldor.org.

Pre-history: Scratchpad II

The Aldor project was initiated at IBM Thomas J. Watson Research Center in the 1980s as an outgrowth of the research work in symbolic mathematical computation.

At IBM, the computer algebra group, led by Richard D. Jenks and including Robert S. Sutor, Barry M. Trager, Stephen M. Watt, and others, were developing a research system for computer algebra called Scratchpad II. This system had its own programming language (compiled to Lisp) which was a precursor to Aldor. An early version of this language was described in the paper A Language for Computational Algebra (Jenks and Trager, pp 22-29 Proc. 1981 Symposium on Symbolic and Algebraic Manipulation, ACM Press). The purpose of this language was to allow users to define their own algebraic structures (e.g. rings, fields, modules...) in a general way. The language provided run-time generics (templates) and used type categories (interfaces) to restrict type parameters to satisfy interface requirements. The language had several limitations and irregularities (e.g. one could write expressions for which there was no well-defined type in the language) and there was not a complete implementation.

A New Language: A#

An effort was initiated in 1984 to redesign the Scratchpad II extension language and provide a complete implementation, sufficiently robust to support a large, complex system. This was led by Stephen Watt, and a first, Lisp-based, implementation involved contributions from

In 1990 Stephen Watt initiated a second, C-based, implementation under the name A# (A Sharp).

A First Deployment of A# in AXIOM

In 1991, Scratchpad II was adopted by the Numerical Algorithms Group Ltd, a not-for-profit UK corporation. It was released by NAG under the product name Axiom.

Work continued on A# for inclusion in the second (1994) release of Axiom. This work was described in the paper A First Report on the A# Compiler (Watt, Broadbery, Dooley, Iglio, Steinbach and Sutor, Proc. International Symposium on Symbolic and Algebraic Computation (ISSAC 1994), pp 25-31, ACM Press). This implementation first became publicly available at the end of 1994 as part of Axiom release 2.

Though this effort was led by Stephen Watt, many people directly contributed to the compiler implementation at IBM, including

Others contributed to the initial mathematical libraries, the documentation, porting and testing, including

A number of executives and managers in both companies devoted many hours to the transfer of A# from IBM to NAG. Thanks are due in particular to

What's in a Name?   A#   to   AXIOM-XL   to   Aldor

For trade-mark reasons, it was decided at IBM that the name "A#" could not be used. Therefore, in its first release with Axiom 2.0, the new language was described as "the Axiom Extension Language", or Axiom XL. Subsequently NAG established the product name Aldor for the language and its compiler.

Use and Development in the Late 1990s: FRISCO and other projects

INRIA (Institut National de Recherche en Informatique et en Automatique) and NAG, together with the French CNRS and the Universities of Pisa and Cantabria launched the ESPRIT Long Term Research Project FRISCO (FRamework for Integrated Symbolic-numeric COmputation) to further research and development in compiler technologies (Aldor-based) and algebraic algorithms. Later, a group at the University of Western Ontario joined this effort. A number of indiviuals enhanced the Aldor compiler and wrote significant Aldor libraries under the auspices of FRISCO and the IBM/NAG relationship, including
At the same time, a number of other individuals made significant contributions to the Aldor world, including
These efforts led to the creation of some significant packages, including

The FRISCO project ended in 1999, but the indivuals who had been contributing to Aldor during this period continued to do so. Participants at the University of Western Ontario wish to acknowledge support from the Canadian Natural Sciences and Engineering Research Council, the Ontario Research and Development Challenge Fund, and the Fields Institute.
 
Subsequently, a number of additional individuals have made significant contributions to Aldor, including

The Present Day

In 2001, NAG ceased distribution of Axiom (and hence Aldor).

Aldor.org was formed as an organization involving NAG and the principal supporters of Aldor to distribute it freely. Work on Aldor now continues:

Aldor is used in projects at NAG, IBM and Seagate as well as a number of universities.

 

There have been many others, not listed here, who have helped in various ways and to them we offer our most sincere gratitude.