I'm not expressing a preference here; this isn't a problem I have. But I do also suspect that a project with as high a profile as curl sees a lot of trivial PRs with less intent to improve the software, than to have the merge show up on the contributor's behalf in Github's various recruiter-friendly metrics.
How that intent plays with Git as Git is done, versus how Github does it, I have no idea. But the existence of the article, and the perspective toward which it seems to be written, suggest to me the answer is "not well" and that Stenberg et al may be seeing the same questions and complaints frequently recurring about this.
But that's my point, what are these metrics that supposedly don't show the contributions anymore? Because the (IMHO) important ones all still show it, because they go by commit authorship in the repo, so you still get listed in the important places and can point to them.
How that intent plays with Git as Git is done, versus how Github does it, I have no idea. But the existence of the article, and the perspective toward which it seems to be written, suggest to me the answer is "not well" and that Stenberg et al may be seeing the same questions and complaints frequently recurring about this.