The author's argues that the 50-50 model of chess is less true and a dead-end.
If this is a matter of which model to use then fair enough, you may wish to use a more exact model than the 50-50 model; but the author is not arguing about which model to use, they are arguing about which model to pursue building upon.
Building upon any model of interest is not a dead end (because it is being built upon!). Even if the underlying principles of the model of interest need to be changed to accomplish something else in the future, it is still useful to develop the model. Approximate truths can also have deep meaning and are sometimes even more generalizable to multiple areas of reality than exact answers. Approximations are no less true than trying to be exact, they are just saying a different thing. Neither is inferior to the other, or at least if exactness really is better than approximation, this is not a good argument for it.
Another commenter pointed out that some models need to be thrown out in order to make room for the new (e.g. the earth-centric view of the solar system had to go at some point), and I think that's valid and hard to argue against; and it seems to align with what the author is saying. But the work done upon the old models was certainly not worth nothing. For one thing, the work done upon the old models is what made the new work possible. I think perhaps the author's issue is that they do not acknowledge that the 50-50 model of chess has value.
p.s. to the author if they read the comments: I actually enjoyed reading your thoughts even if I disagree with them.
Tbh, I didn't think he gave that message. If anything, he even identifies the 50-50 model as a local "maximum" (i.e. it has value). To me the main point was to have a mindset that is willing to challenge the possibility of whether such an optimum could actually simply be a local one.
If this is a matter of which model to use then fair enough, you may wish to use a more exact model than the 50-50 model; but the author is not arguing about which model to use, they are arguing about which model to pursue building upon.
Building upon any model of interest is not a dead end (because it is being built upon!). Even if the underlying principles of the model of interest need to be changed to accomplish something else in the future, it is still useful to develop the model. Approximate truths can also have deep meaning and are sometimes even more generalizable to multiple areas of reality than exact answers. Approximations are no less true than trying to be exact, they are just saying a different thing. Neither is inferior to the other, or at least if exactness really is better than approximation, this is not a good argument for it.
Another commenter pointed out that some models need to be thrown out in order to make room for the new (e.g. the earth-centric view of the solar system had to go at some point), and I think that's valid and hard to argue against; and it seems to align with what the author is saying. But the work done upon the old models was certainly not worth nothing. For one thing, the work done upon the old models is what made the new work possible. I think perhaps the author's issue is that they do not acknowledge that the 50-50 model of chess has value.
p.s. to the author if they read the comments: I actually enjoyed reading your thoughts even if I disagree with them.