The Experiment Markup Language (ExptML) is an approach to digitally capture science as it happens. Given the advances in computation, online databases, and the semantic web it is time to bring the fundamental process of doing science into the 21st century. While researchers at institutions, government agencies, and industrial facilities do a large amount of work electronically there are still two areas were they are inefficient: collection of raw science data (e.g. in a laboratory notebook) and integration of digital data into the laboratory notebook.
ExptML aims to solve these major issues by providing a mechanism to store the data and metadata about the scientific process as it is done. This approach breaks down research into the fundamental activities that are undertaken by scientists and attempts to define datatypes (see left) that capture the data being generated and the context (via metadata) of how it was achieved. With this approach not only can the data be search but also the context around the data (i.e. its semantic nature) providing for a much more efficient search, which is needed considering the large amount of data generated.
ExptML is not monolithic. While the basic premise is to provide a complete record of scientific research, the idea that ExptML alone can represent everything in science is unrealistic. There are a wide variety of other markup languages, ontologies, and controlled vocabularies either already developed or currently under development that take advantage of domain specific expertise for those areas. ExptML is built (and will continue to evolve) in a way to incorporate these other technologies wherever it makes sense to do so. See the datatype pages for some examples of technologies that have already been incorporated into ExptML.
Individual schema and example files are available on the datatype pages, and a complete set of current schema is here.