In 1983, I told Richard Manning about Gale and Shapley's beautiful 1962 paper on matching. He asked whether in the Gale-Shapley it was in the interest of all participants to tell the truth. We rather quickly showed that in general it is not in the interest of the recipients of offers to be truthful. In fact we were able to show that no mechanism can guarantee efficient assignments and be cheatproof. We were very pleased. We sent it to a journal, only to learn that Al Roth had beat us to it in a paper that was to appear in JET in 1984. Roth's paper was not only earlier, but deeper and better than ours. But our proof is different from his and our example might still be of interest.