It is now thirty years ago that the concept of stable model has been proposed by Gelfond and Lifschitz in a seminal paper at ICLP, in a joint meeting with the Symposium on Logic Programming (SLP), at a time when non-monotonic reasoning and logic programming were among the hottest issues discussed in Knowledge Representation and Artificial Intelligence. While the interest in non-monotonic logics and formalisms has decreased over the years in general, the proposal to use logic programs for declarative problem solving, coined answer set programming, has led to a stream of research in which, in mutual support, theory and practice of logic-based problem solving have led to a flourishing branch of computational logic. This talk will make a tour through the history of Answer Set Programming, take a look at the current state of research, and speculate about possible futures of the paradigm -- of which multiple exist. It will consider developments in perspective and address some issues and lessons to be learned, which may be kept in mind to experience longevity and leave the opportunity to celebrate in decades from now again.