From: Dominic Benson <[log in to unmask]> Date: Fri, 24 Feb 2017 10:45:23 +0000 It's a similar situation to when one changes subscription agent. Payments from the new agent are transacted and reconciled but, after the grace period expires at perhaps a handful of publishers, the apparent "non-renewal" by the former agent may trigger a cancellation process and subsequent loss of access to subscribed content. The supply chain breaks too easily. Similar events happen when titles transfer between publishers as Ross has noted: http://rossmounce.co.uk/2015/03/07/another-day-another-elsevier-website-illegally-selling-articles/ Imagine what might happen when a publisher's platform migrates to a new service. There have been several instances in recent months. Check (again), please? Kind regards, Dom Benson E-resources Librarian, Library, Information Services Brunel University London -----Original Message----- From: Rick Anderson <[log in to unmask]> Date: Thu, 23 Feb 2017 17:03:32 +0000 It would be very interesting to see a rigorous, data-driven study of the extent of this problem. Ross has found a handful of articles that aren’t being made OA despite an APC being paid, and presumably there must be more – but are there ten more, or a thousand more? I wonder if you could arrive at a valid conclusion through a sample-based study: take, say, ten issues each from 20 or 30 hybrid journals from a variety of publishers, and see how many of the putatively OA articles in them are behind paywalls. (But how could you know for certain whether an APC had been paid for any particular article? Hmmm.) The characterizations and inferences in Ross’s piece strike me as a bit over the top – but clearly there is a problem. I’d love to get a better idea of whether it’s small, medium-sized, or large. --- Rick Anderson Assoc. Dean for Collections & Scholarly Communication Marriott Library, University of Utah Desk: (801) 587-9989 Cell: (801) 721-1687 [log in to unmask]