I think both methods are valid. One tutorial compliments the other. They do however, start in different places. The first does not address DSLR_RAW conversion, while the second does. There are two approaches to the problem and for one or the other according to preference both are acceptable. The developers have gone into print in relation to the reasons for DSLR_RAW conversion during image calibration and that is the basis of the DSLR_RAW workflow.
I think some image sets are problematic, which doesn't warrant changing either approach, or choosing one over the other in my view.
Cheers,
Rowland.