Eric, what would that "few million for a good study" buy? Might it not be wiser spent on a full-blown market test, using software prototypes? That is exactly the kind of thing I meant. Several smart-card payment systems have been deployed in medium size cities in Europe as trials to see just how much they'd cost in practice to deploy. These trials cost more than just a few million, but prototype implementations of each of online and offline sorts of systems, complete with standard marketing tools such as focus groups and limited scale deployment, as, for example, inside an amusement park. Whatever the actual figures are, there are too many of them now to each side's benefit to say definitively what will be the best in any particular market segment, even if some of the choices are clearer than others. Eric