The cite I provided is taken from a question-and-answer format in the book itself, with Albert Einstein answering the questions posed by Infeld, dr_lew.
I provided a link to archive.org. The entire book is available there, for free download. Give it a look.
Boy, I'm not seeing this. I find your excerpt on page 212 of the Google Books "review" and there is no such Q&A evident anywhere leading up to it. There are some sequences of rhetorical Q&A and one where a "classical physicist" is imagined to be giving responses to questions, but I see no instance of such an interplay represented as being between the co-authors. So, I think my caveat applies.
Anyway, let me address the point itself. It comes from a section emphasizing the difficulty of identifying an inertial frame, as posited by Newtonian dynamics, and makes the statement then, that this is not necessary to do.
But what do you do in this case?
"Nothing is more distressing on first contact with the idea of "curved spacetime" than the fear that every simple means of measurement has lost its power in this unfamiliar context" - MTW, GRAVITATION
This is on page 5 of the introductory chapter 1, prefatory to a description of "Locally Lorentz" frames of reference, as determined operationally by a "Lorentzometer", which is pictured ( a sort of Rube Goldberg affair. ) The point is that the GLOBAL inertial frames of Newtonian dynamics have been abandoned, but the idea of inertial frames ( i.e Locally Lorentz frames ) has not. Locally Lorentz frames are the bedrock of GR, by which everything else is triangulated.
In contrast to Newton, gravity is not considered a force in GR, and any body in "free fall", e.g. in orbit around the earth, will pass the Lorentzometer test. The geodesics of spacetime are simply those paths followed by objects in free fall, and we can quickly see that the planetary orbits are such geodesics.
Of course, this determination, made operationally as described, is independent of "choice of coordinate system". This amounts to a more sophisticated expression of Copernicanism, and while it may sweep away the TERMS in which Copernicanism was stated, it does not in any way vitiate its substance.