While computational experiments have become an integral part of the scientific
method, it is still a challenge to repeat such experiments, because often,
computational experiments require specific hardware, non-trivial software
installation, and complex manipulations to obtain results.  To address this
problem, we are building infrastructure to support the life-cycle of
'reproducible publications': their creation, review and re-use. This
infrastructure,  which builds upon and extends the open-source VisTrails system,
makes it easier to generate and share repeatable results by making provenance a
central component in scientific exploration, and the conduit for integrating
data acquisition, derivation, and analysis as executable components throughout
the publication process.   Besides giving authors the ability to link results to
their provenance, it enables reviewers to assess the correctness and the
relevance of the experimental results described in a submitted paper; and, upon
publication, it also allows readers to repeat and utilize the computations
embedded in the papers.  In this talk, we give an overview of this
infrastructure and present case studies that show how this infrastructure has
been used. We also discuss its limitations and open problems that require
further research.