For a new technology to be successful, it has to come at the right time. Waiting too long before developing an essential module means that ad-hoc solutions will be deployed, which causes backwards-compatibility problems and high costs for the people who have to convert things to the new standards once they arive. Developing something too early may mean that it will be forgotten by the time it is needed, or worse, that it can't be used, because the intermediate steps have turned out to be different than what was expected.
But what is the right time? Hard to say. But some of the symptoms of its arrival are that people start asking for something (although it is quite an art to recognize what they are asking exactly), that there are people available with ideas, and especially that the solution appears obvious.