I second the opinion of Shawn. EBSCO, PROQUEST, CABI etc still maintain good evaluation criteria and it is not easy to be included. OK, I do agree that some low quality journals are there. But still the overall quality is OK. Exception is not the rule. For some exceptions, you can not blame the entire system. These low quality journals can be from Open access journals or from subscription based journal. Therefore, if you are speaking about the faults of EBSCO, you should discuss both OA as well as subscription based journals. It will complete the discussion.
I expect more patience when talking about others’ small error. Please refer to the entirety of the criterion/error from both side. It will give you more credibility. You should remember that nobody is perfect. And your list is also not perfect and your evaluation criteria are also not perfect (Reference you have deleted many silly or laughable reasons from your version 1 criteria (http://scholarlyoa.com/2012/08/04/criteria-for-determining-predatory-open-access-publishers/) and improved it in version 2 (http://scholarlyoa.com/2012/11/30/criteria-for-determining-predatory-open-access-publishers-2nd-edition/)). Therefore, it will not be justified to cancel all your hard work by showing 4-5 borderline publishers, who were dropped previously from your version 1 list or in future some more names may be dropped if they show improvement in publishing practice.
As Elsevier published 6-8 fake journals, you can not blame that total Elsevier quality went down or Elsevier is promoting fake journals (Reference: Elsevier published 6 fake journals: Source: http://www.the-scientist.com/?articles.view/articleNo/27383/title/Elsevier-published-6-fake-journals/ or http://en.wikipedia.org/wiki/Elsevier). Most of these AI agencies are good and maintain quality.
