The article itself mentions scalability issues; but beyond this - there are often many complications in Makefile construction which result from idiosyncracies o

The article itself mentions scalability issues; but beyond this - there are often many complications in Makefile construction which result from idiosyncracies of libraries, utilities, platform settings, tool versions, etc. etc. I used to write Makefiles - basically in the vein of this post's suggestions; but for several years have switched to using CMake. I have not had reason to look back. Why would I want to write Makefiles directly? What's the benefit?