Squeezing as much profit as possible for shareholders while externalising costs. It’s a poster child when you think about it!

The copies on the author's pages and in open repositories are obliged to be pre-publication drafts, so you cannot cite them.

In fact, under control of large publishers journals have become fairly inventive about making pre-publication manuscripts unusable. For instance, I recently published in a prestigious journal that had an online editing system so that all final revisions stayed in the system - no author copy of any kind, every proof marked with author name and large watermarks -, and they made lots of small, sometimes unnecessary changes in the very last editing step. They also don't paginate the final versions that are published online first, so you can only know the real final pagination a year or so after the first version has been published online.

A bearded man calls this "surplus". Scientific journal business is the pinnacle of capitalism, what every company strives to become, including the ones fellow HN people here work at and start up.

The designation that combinations of words are a form of property that can be owned, which then become assets naturally funneled towards those interested in playing the capital meta game and away from those doing the actual work.

To a scientist, the ability to copyright their paper is a vulnerability. Get rid of the concept of imaginary property and there is a lot less for Elsevier to demand from them.