Your analogies are completely off the mark. It is not the same at all. The cornerstone of science is that other people can reproduce your results. Period. There is no use publishing otherwise. Withholding key information, which is so ubiquitous now, is a great disservice to the scientific community.
The question is not whether reproducibility is good, it's how much labs should invest upfront in producing descriptions of protocols. My argument is that they should probably invest more than they do now, but not enough that pharmacutical companies are able to reproduce a given experiment without talking to the lab.
Science is a collaborative process. There's nothing wrong with collaboration being part of the reproducibility process, as long as the person doing the reproduction maintains their objectivity.
People often misunderstand the idea of reproducibility in science. The idea is not that scientists need to given a complete, easy to follow set of procedures that will let anyone reproduce an observation. In some areas this is close to impossible. For example, in high energy physics there is only one equipment in the world that can reproduce (with luck) some key experiment. The idea is that, with enough effort and funding somebody else could possibly reach the same conclusions. If you don't agree with some published results, there is an easy way to do it: create your own experiment and publish the results. If the result conflicts, then a new step is to determine why and in which conditions there is a conflict. Science evolves through the debate of ideas and observations, not because somebody is sharing a cookie-cut recipe.
