This post originally appeared on the Software Carpentry website.
From Ana Nelson, a neat new reproducible research tool called Dexy that takes the pipe-and-filter model to new heights (or extremes, depending on your point of view :-) ). There's a quick intro here, and a video of her recent talk about Dexy here. I'm already thinking about ways of combining this with consistency-checking tools so that if the descriptions of results get out of step with the results themselves, alarms will go off. Check it out, and let us know what you think—is it useful and approachable enough to include in this course?