I have followed this discussion without really getting an idea of what Don wants. Here are a few random thoughts.
For expressing mathematical notation, LaTeX works fine, and is used for most mathematical journals. Executing mathematical notation without change is a challenge. The only language I know to seriously address this is Fortress, e.g. http://research.sun.com/projects/plrg/PLDITutorialSlides9Jun2006.pdf It is not clear to me whether this is desirable, given the ambiguities and context dependencies in mathematical notation. Attempts to make notation more precise are dismissed as having too many computer science trappings. Languages such as Maple, Mathematica and Matlab have included such trappings without deterring users, and it is not a huge stretch to guess what Mathematica might mean by Integrate[x^2,x]. I don't see why explicit definitions are excluded: "Let f(x,y)=x^2+y^2" is a perfectly normal part of mathematics. What would help me the most would be some extended examples of what is expected of the system. At the moment I cannot see why a calculator is not sufficient. Best wishes, John ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
