The Open Provenance Model

Provenance is well understood in the context of art or digital libaries, where it respectively refers to the documented history of an art object, or the documentation of processes in a digital object’s life cycle [3]. Interest for provenance in the “e-science community” [9] is also growing, since provenance is perceived as a crucial component of workflow systems [2] that can help scientists ensure reproducibility of their scientific analyses and processes. Against this background, the International Provenance and Annotation Workshop (IPAW’06), held on May 3-5, 2006 in Chicago, involved some 50 participants interested in the issues of data provenance, process documentation, data derivation, and data annotation [5, 1]. During a session on provenance standardization, a consensus began to emerge, whereby the provenance research community needed to understand better the capabilities of the different systems, the representations they used for provenance, their similarities, their differences, and the rationale that motivated their designs. Hence, the first Provenance Challenge was born, and from the outset, the challenge was set up to be informative rather than competitive. The first Provenance Challenge was set up in order to provide a forum for the community to understand the capabilities of different provenance systems and the expressiveness of their provenance representations. Participants simulated or ran a Functional Magnetic Resonance Imaging workflow, from which they implemented and executed a pre-identified set of “provenance queries”. Sixteen teams responded to the challenge, and reported their experience in a journal special issue [6]. The first Provenance Challenge was followed by the second Provenance Challenge, aiming at establishing inter-operability of systems, by exchanging provenance information. Thirteen teams [8] responded to this second challenge. Discussions indicated that there was substantial agreement on a core representation of provenance. As a result, in a workshop on August 7-8 in Salt Lake City, the authors met, and crafted and iterated a data model, which is presented in this paper.