Fascinating article, but I have to dispute the idea that the quest for elegance and simplicity is "wishful thinking".

Let's start with a very basic idea - the whole reason to come up with an explanation is that it is a lot simpler and therefore more elegant than the data it explains.

If we didn't care about simplicity, we could "explain" the universe just by pointing to it. For an explanation to "work", we have to be able to understand it!

While essentially unprovable, you can make good statistical arguments that if you are given two models of the same system, and one is "smaller", it is the better model - in other words, Ockham's Razor.

And as you point out, in the past our quest for simplicity has worked out really well in the past. In Newton's equation for gravitational force, the attraction decreases as the square of the distance. In the 400 years since Newton, we've measured that "squared" exponent to greater and greater precision, but it seems to be "exactly" 2.

Yes, physics now has a lot of theories like string theory that don't seem to make testable predictions. But everyone agrees that a "theory" which doesn't make testable predictions isn't actually a theory at all. People propose these theories in the hope that it will lead to some sort of testable result we can perform an experiment to measure.

Researchers didn't set out to make string theory or other unification theories to be annoying - the trouble is that we have two "perfect" theories, Relativity at the macro like and Quantum Mechanics and the micro level. Both have been heavily tested in their domains without flaw, and yet they contradict each other.

So science is struggling to reconcile these contradictions.

What would this ugly theory look like, anyway? A large collection of disconnected facts and equations that had strong predictive value but nothing else? Maybe that's all we get, but I for one would be disappointed.