But many markets (the labor market, the health care market, etc) work in such a way that it's ridiculous to assume that most participants know all public and private information all or most of the time.
Also what people mean when they say markets are efficient in other contexts means something very different. The EMH specifically describes how quickly and what kinds of information get incorporated into asset prices. It doesn't make claims about how markets organize production or optimize utility without centralized direction, which is usually what people mean when they say free markets are efficient.
I think a lot of confusion has arisen from the very general-sounding name of the hypothesis which does not reflect its relatively narrow claims. Not that I buy the EMH (no pun intended), but that's a different story.
Another requirement is that entrepreneurs have fair market access to capital. This assumption is true in many developed countries, but not in many developing countries where you need connections to obtain a loan. But even in developed countries, it's difficult to get extremely large loans on the merits of the financials alone, which explains why large infrastructure projects like highways, airports, and power-plants often don't get built solely by the private sector.
Also EMH, and that's the point Fama uses a lot to battle critics, does not put a time frame on when the price adjustment happens.