...but you knew that already
Gina Kolata at the NYTimes has an interesting article about discovering the already discovered. People working on new mathematical algorithms might publish and name an improvement, only to find out later that the math had been done under a different name years earlier. Worse, small scale drug trials, showing benefits of a drug after surgery, could be needlessly repeated (including placebo controls) because the work was not published.
Final word goes to Larry Shepp: "Yes, but when I discovered it, it stayed discovered."