There are plenty of scientists out there who poach free online data sets and mine them for additional findings.
And this is a good thing, despite your word "poach". Analyses which would not have occurred to the original experimenters get done, and we get more science for our money. For many big data projects (e.g. the human genome project, astronomical sky surveys), giving 'poaching' opportunities is the primary purpose of the project.
A former boss of mine once, when reviewing a paper, sent a response which was something like this:
"This paper should absolutely be published. The analysis is completely wrong, but it is a wonderful data set, and somebody will quickly publish a correct analysis once the data is available."
Now I need to stop wasting time on /. and return to my work in hand, which, as it happens, is 'poaching' data from
Ingman, M., H. Kaessmann, S. Paabo, and U. Gyllenstern. 2000.
Mitochondrial genome variation and the origin of modern humans. Nature 408:708--713.