[gampleman]

I think the main point of the article is that you can use property based testing not as a software quality tool, but as a very quick way to verify/disprove your assumptions. The MapReduce stuff is just a simple enough example to illustrate the point.

I've used property tests the exact same way many times. In complex algorithms assumptions can be hard to think through, but a property testing tool can find counterexamples shockingly quickly.