Handheld computers have been criticized as one of the most excessively hyped new IT products of all time. This paper looks at handheld computing predictions made over a 10 year period, investigating what went wrong, and what went right, with handheld computing predictions. Handheld computing predictions can be divided into three phases, depending on the product concept definition widely held at the time: handheld computers as penbased computers, personal digital assistants, or handheld companions. While longer-term predictions were inflated in the first stage, they were surprisingly accurate in the second stage and excessively conservative in the third stage. The complaints about over enthusiasm and hype have more to do with incorrect product concept assumptions than poor guesses about the size of markets— technology directions are just as difficult to predict, or even more difficult, than technology sales.