Many thanks to everyone - you have given us new determination to follow the MathML route. I now have a functional MathML engine which will soon be capable of managing most of the common chemical mathematical operations (e.g. at first year undergraduate level). It is also capable of symbolic differentation of many of the functions. We are writing a paper on "the mathematical basis of semantic physical science" (with examples based on molecular energies from so-called forcefields (actually energy fields). See http://en.wikipedia.org/wiki/AMBERfor a useful account. Our approach now involves only generic engines (i.e. MathML processors or standard tools such as Conjugate Gradient minimisers) and most of the complexity of current programs has disappeared in the declarative approach. The chemical functionality is also largely declarative (in CML). We'd be very grateful for comments on the draft (in perhaps a few days). If this approach is useful, then people are welcome to our (fairly simple, Open Source, XOM-based) engines for processing MathML. -- Peter Murray-Rust Reader in Molecular Informatics Unilever Centre, Dep. Of Chemistry University of Cambridge CB2 1EW, UK +44-1223-763069
Received on Friday, 16 March 2012 17:25:58 UTC