The Equation Map is an idea for a project that currently I will never get to pursue, due to the size of the undertaking and the time constraints it would require. None the less, it's something that will eventually be done should humans continue to progress. I view it primarily in 2 parts. Part 1 is the math side and part 2 is the science side, starting with physics.
The math side recognizes that based on the 4 basic operations of math, adding, subtraction, multiplication, and division, and a small handful of other basic operators, exponentiation, summation, product functions, derivation, integration, and matrix operations, that a large portion, "99%," of all possible math formulae can be systematically created using combinatorics of those operations. Such a system would provide all possible formulae arranged by complexity, and would serve as an "Equation Map" of all equations. This would be helpful such that when an equation was needed for a model, it would provide a place to start looking for it, or from which to branch out.
The physics and science side begins with the current known given equations of physics, and clusters and links them together based on variable dependencies. For example, all equations that used the variable c for the speed of light would be linked, or all equations that used meters such as velocity would be linked, and so on. From there, one would be able to investigate which variables were truly independent versus dependent, simplify formulae, as well as to start analyzing the overall structure of those formulae clusters. Those groupings of the physics map could then be compared and aligned to the math equation map, resulting in a smaller divide between pure math and physical math applications. This would also facilitate and increase the speed at which unsolved physics and science questions would be solved.
This concept aligns with the Parameter Composite Modeling portion of the Rogue Physics page and is roughly the infinite parameterization to which it refers.