Editors-in-chief, and what software lets them do
Pyramid scheme by Ilana Gershon (h/t Siderea).
This post raises interesting issues, independently of what you think about the discipline at hand, the distracting lingo and wordiness, and the anti-capitalism dragged in by its head and shoulders. It's telling a story of an OA journal ("HAU") ruined as a (most likely unintended) consequence of the software platform (Open Journals Systems) placing excessive control in the hands of the editor-in-chief. In particular, she mentions that the platform is quite efficient at splitting the work into independent bite-sized pieces that can be handled by different people more or less automatically, while automatically taking care of communication; but this very automatism means that no one but the editor-in-chief gets to see how the sausage is made. (Sorry if I am misstating the point; either way there are several other points as well.)
Mathematicians, of course, will remember another story of an editor-in-chief destroying his newly-"liberated" journal while leaving the archive in legal (copyright) limbo. Lessons obviously have been learnt in terms of licensing; but perhaps it's worth separately thinking about how to avoid accidentally concentrating paper in one person's hands as well.
Coincidentally I had a friend visiting me, and somehow our conversation steered to one of the mysteries of modern mathematics: how it comes that Duke Mathematical Journal, nominally a general-purpose maths periodical, tends to only accept papers from a few sub-disciplines, while other parts of mathematics remain significantly underexposed (various journals are biased, but few as much as Duke). My friend mentioned that it might be due to the important role of the managing editor, who in Duke gets to make all the acceptance decisions (while at other periodicals, the decision is usually delegated to the editor closest to the subject of the paper). Now, Duke is perhaps the opposite of an OA journal; the root cause seems remarkably similar.