Chart stars are ubiquitous in old media (which more people still understand than new media) and they are pre-selected by the commercial music process to be tole

Chart stars are ubiquitous in old media (which more people still understand than new media) and they are pre-selected by the commercial music process to be tolerable to a broad base. I think the main costs of switching away from chart stars are informational. To a lesser extent, the cost of learning new technology. To a greater, the cost of finding anything of interest. Those costs shouldn't be underestimated. Finding gold amidst dross (even for subjective values of gold) is a Bayesian evidential process costing nonzero thermodynamic work. In other words, it is never going to be a free lunch. And right now it is very expensive. The charts are reliably more likely to carry something you'll tolerate, even if there are things out there, undiscovered, that you'd properly enjoy. So fair enough, for now, the charts continue. But this is an informational cost, and if one thing has become clear of late it's that premising your industry on obscurity is a mug's game. Youtube and friends are indeed pouring machine learning technology into uniting you with your own personal chart; I expect a crossover moment will happen sometime in the next ten years after which you will be able to rely on getting better hits off a taste-learning algorithm than off the radio. And then we will really see if the masses want Rihanna.