Starting to make a new language, a new library, a new app, is easy and exciting. Creating the necessary support for efficient and economical software engineering, growing an industry to back it up, writing books and tutorials, and engaging a community that will maintain it in the long run is boring and tedious and very, very hard.
As these matters become harder, we starting a new one without the restrictions and the baggage of past mistakes is quicker than getting people mobilized towards a common goal.
So, there'll be more abandoned languages and libraries and apps that nobody uses. And it'll never stop.
It's similar to why we keep our garage filled with sports equipment we never use, guitars we'll never play, and endless unfinished projects. The beginning is easy and exciting and filled with possibilities. Getting through the finish line... that's what counts.