Oddly, I interpreted "objectively," as necessarily immeasurable, in contrast to commenters here who state the prerequisite is that "good," be measurable. The example I think of is judged sports, which are essentially arts with a measurement criteria bolted on to them after the fact to facilitate governance. In these cases, the interpretation of the performance does more to legitimize the governance and the judges than it does to meaningfully evaluate the performance itself.<p>In this sense, I'd posit "objectively good," is a question of beauty, and not measurement criteria. Those sports suffer from the Goodhart's Law problem (<a href="https://en.wikipedia.org/wiki/Goodhart%27s_law" rel="nofollow">https://en.wikipedia.org/wiki/Goodhart%27s_law</a>), where you don't need a competition to recognize someone is objectively good at something if they perform it beautifully.<p>The idea of "good," as a set of measurable scalar quantities presumes that you can express the object in those terms, which in the case of writing or a performance, you can't. There is no set of instructions that can reproduce a "real" performance.<p>You can produce something where people won't care about its difference from the "real" one (lots of economics and cog.sci on that one), but the existence of a simulation does not change the fact of the existence of the real.<p>For example, if some future GPT-7 produced the literary equivalent to crack cocaine, it would still not be Ted Chiang, whose work is beautiful and could be said to be objectively good. People might prefer this new crack-lit, which mutes their ability to sense ugliness, parasitism, disgust, or horror, but its existence does not obviate the existence of Chiang. That essential existence is what makes Chiang an objective phenomenon, and the beauty of his work is what makes it good.<p>I don't think there are short answers to this question though. :)