Reproducible Research

In a nutshell:

"An article about computational science in a scientific publication is not the scholarship itself, it is merely advertising of the scholarship. The actual scholarship is the complete software development environment and the complete set of instructions which generated the figures." -- D. Donoho

As an example, at university I wrote a paper on network coding and referenced the Microsoft Avalanche project (http://research.microsoft.com/en-us/projects/avalanche/). What always puzzled me was that they never published source code to back up their claims to fame. Why? Maybe I'm being naive, but to me the openness of the scientific community implies that you must provide other researchers the opportunity to reproduce your results.

If this were a paper on chemistry you'd describe the volumes, concentrations, and proportions of all substances used. So why don't computational science papers offer source code and instructions on setting up the build environment? What am I missing?