There are two different types of technological development. The simpler, and simpler to predict, is refinement. You take something that already exists, and you make it a little better in some way. What that “better” is depends on what the original tech was. For an MP3 player, making it smaller might be a good thing. For a car, maybe not, since you still need to fit people inside it. For the car, you might make it more fuel-efficient, or you might give it a greater top speed, or make it safer, or re-arrange the cabin to make it more space-efficient, or any number of other improvements. The MP3 player would be improved by giving it a better interface, or a longer battery life, or making it more durable, or giving it more capacity. And we can expect that all of these incremental improvements will be made, to some degree, as time goes on. So the car of the future will be faster, safer, and more efficient than the car of today, and the MP3 player of the future will last longer on a charge, store more songs, and be easier to use.
But there’s a limit to how far you can refine any given technology, and that’s where the second type of advance comes in. The second type of advance is innovation, making a completely new product. This new product may fill a similar niche to an existing product, but it does it in a very different way. It might start off not as good as the older style product, since it hasn’t had the years of refinement yet. But it will eventually be refined as well, and eventually hopefully supass the old style. One example of this might be electric cars: An electric car is not just a slight modification of a design of a gas car; it’s a completely different sort of thing. And right now, electric cars are worse than gasoline ones in almost every way: They’re slower, they have a shorter range, they’re more expensive both to buy and to maintain. But they have a different set of limitations than gasoline cars, and might, with sufficient refinement, end up better than them. The thing is, while incremental advances are somewhat predictable, innovations are not, in general. They mostly depend on someone getting a good idea, and if nobody happens to have gotten that idea yet, they don’t happen.
Of course, there’s some overlap. For instance, the fuel injector, which replaced the carberateur, was an innovation, but it resulted in an incremental advance in the automobile. It was easy to say that somebody would come up with a way to make a car better; that’s the predictable type of advance. It was more difficult to predict that what they came up with would be a better way of mixing the fuel and air, since that specifically is an innovation.