Standards efforts

Reproducibility of results is an important part of any kind of scientific study. The ability to easily share computational models and data is crucial to this. However easy it is to make copies of digital data, the utility of this data is often limited by its binding to specific softwares.

For example, in case of computational modeling, the models are usually developed for specific software environments and may require significant amount of time and technical expertise to port to another system.

As long as the underlying science remains unchanged, a good model should give the same results on all good software systems.  Also, good models can be looked upon as assets of the community which can be used for further research as well as for composing larger models. Beyond research, well-established computational models are invaluable for education and hence easy access to them is very important.

Similarly, if some experimental data is in a software-specific format, it is impossible for someone without access to the details of that format to interpret or analyze that data.

Common standard formats for defining models or for specifying experimental data can address these issues. INNNI is involved in several community-efforts to define such standards: