Eugen Leitl <eu...@leitl.org> writes: > It seems to be in everyone's interest to allow easy interoperation > between the toolkits. However, in my experience it is a frustrating and > difficult task, even for relatively simple tasks.
True. > I believe that a common set of core modules is required. The most Not necessarily, although it might be the most practical approach. The minimum requirement in Python is a common core interface, implementations could be different. > and optimised for specialised work. If a common molecule definition can > be agreed upon by the major authors of the toolkits concerned, > interoperability will be made MUCH easier. Other multipurpose molecular Definitely. The problem is that this is a major effort, first for defining a sufficiently universal set of classes, and then for adapting all the codes to it, plus eventually providing a compatibility layer to keep old client code working. It won't happen overnight. > The question is whether the will exists to do this? Many of the toolkits On my side, yes. Konrad. -- ------------------------------------------------------------------------------- Konrad Hinsen | E-Mail: hin...@cnrs-orleans.fr Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.56.24 Rue Charles Sadron | Fax: +33-2.38.63.15.17 45071 Orleans Cedex 2 | Deutsch/Esperanto/English/ France | Nederlands/Francais -------------------------------------------------------------------------------
