This is true, but as libraries increasingly move from a "just in case" to "just in time" model of collection building, probably not relevant. We want access to the book when it is needed, which is fine if it is already open, but less helpful if we have to wait until enough people subscribe. And we probably only want to pay for that book, not a whole bunch of others that might be from the same press but aren't necessarily of interest to our community.

I think this is the crucial blocker at the moment. Libraries try not to buy individual books from lots of individual publishers - we buy them from aggregators who can deal with the scale. Equally, if we buy a print book, we order it, and it turns up and we deal with it. Increasingly we buy it shelf ready, so all we have to do is receive it and stick it on the shelf. If its an ebook, we try to buy it on a known platform, to help our users find it and for us to understand usage.

But with OA books, the plethora of sites and pricing models adds a lot of processing and handling costs, even if we have time to understand and track them. Just reading your table makes my head hurt a bit. Without some aggregation, it is all too hard and becomes piecemeal.