Most cultural production in any medium is forgettable. How many composers do you think were alive at the same time as Beethoven? Hundreds? Thousands? How many of them do we still listen to today? Tens, maybe less.
Even more recently, let's take The Beatles and their contemporaries. Of course, The Beatles were not the only seminal artists of their time. They were active at the same time as David Bowie and The Rolling Stones. But again, how many of their contemporaries produced albums that are now forgotten by everyone who listened to them? How much of what's left even now will be gone in another 75 years, when basically everyone who was alive in the 60s is dead?
The only difference between the 60s (or early 1800s) and now is that we haven't yet forgotten all the crap that is being put out today. Will future historians of pop music assign more than a footnote to Katie Perry or Taylor Swift 50 years from now? Will they even get a footnote?
So the reason old art is the best art is simply that we've had time to weed out all the crap. The remaining art that we still remember is the best of what was released at the time. It's not the best because all old was better, it's the best because it's been better curated.
In fact, for all we know, the overall proportion of art that is good might be even better today, but that doesn't change his proposition that we don't really need new art in order to experience great art for the rest of our lives.
In photography, amateurs are all over, with top equipment (it's an old joke that non-pros have the better equipment as they can better afford it). If I go to look at pictures of some place, it may very well be that professional photographers are providing zero additional value to my viewing enjoyment. Why should this be different than any other area?
Statistically speaking, yes, some of them would.
If context counts for a lot then it doesn't matter how good the best of the old stuff was when it was created or how easy it is to figure out which bits are the best today. Newer stuff will still be better now because it's more relevant. Much of it won't be better to people in the future, but that doesn't matter to people now.
Personally, I think there's a ton to be said for context. People want characters and situations they can relate to. At the end of the day, I'd generally prefer to read something where the protagonist uses an iPhone and cooks with a microwave than one where the protagonist writes letters and has his food cooked by slaves.
And of course it doesn't have to be as dramatic as smartphones and slavery. Even a book from just a couple of decades ago is going to feel remarkably different due to changing attitudes towards everything from homosexuality to nuclear war.
But he completely refuses to consider such contextual advantages. He states flat out, "People are better off reading the best books, and the best ones are predominately the ones that already exist, there is more than can be read, and new books have no compelling advantage over the classics." (Emphasis from the author.)
The concluding sentence is an excellent summary of the article, although not the way it's meant to be: "I have started with common-sense grounds and wound up somewhere strange." In my experience, when people justify things with "common sense" they often mean unsubstantiated claims that they don't want to have to justify explicitly. If you start with "common-sense grounds" and end up somewhere strange, consider that maybe your starting assumptions are faulty!