The problem with ascribing error here, when it comes to published copy, is that it is not always the author's fault. As pointed out above, in the old days, text did not go straight from the author's pen (or typewriter) to the published page--neither in the newspaper, magazine, nor book business--nor does it go directly from the author's computer to the printed page today. And in my experience, having written hundreds of magazine articles, it's rare when the magazine returns the edited version to the author for a final look prior to publication. An unnecessary step in most cases, but one which would be relatively simple, in the age of computers and email.
Catching minor errors can become very much a "gotcha" game, and always ascribing them to the author can be a case of unwarranted attribution.
