Title: Everpub: reusable research, 21st century style
Modern research is strongly dependent on computational tools. In biomedical research, computers are indispensible for analyzing large amounts of data and for understanding complex models through simulation. While the rapid advance of computing technology has boosted progress in many fields of science, it has also created barriers for scientific communication, which are particularly painful in the ongoing transition to Open Science. Permitting other researchers, let alone non-specialists, to fully understand complex research requires not only publishing software and datasets, but also providing a computational environment in which readers can explore these methods easily by changing parameters and input data.
Over the last years, a large number of tools have been developed that address specific aspects of this problem, and much valuable experience has been gained from their use. However, orchestrating all these tools for either producing or exploring published research requires a level of technical competence that few scientists have or even wish to acquire. We propose to drive a vertical spike through this problem space, by integrating existing tools and methods into a unified whole whose use requires no more technical expertise than performing computations on a desktop computer. As a consequence, scientists will be able to understand, evaluate, and re-use the work of their peers.
We emphasize that we do not expect to provide a definitive solution to the problem in the near future. Our immediate goal is to provide a usable platform for publishing computer-aided research, which will then be improved over time by bringing onboard scientists, librarians, and publishers. We aim at building a community around these ideas, which is a critical part of the success of the project. In addition to enabling Open Science, this project is also meant to be an example of Open Science.