This is, à mon avis, completely nuts for several reasons.
First, the idea of forced obsolescence is probably so repugnant to most librarians that, rather than buy and re-buy the same e-book over and over, they're probably just not going to purchase HarperCollins e-books.
The publisher seems to be betting on librarians' collective fear that if they don't stock their e-books, patrons will simply buy e-books on-line and the libraries won't be able to justify their continued existence. If there's a way to breed ill will among your customers (that is, the librarians actually purchasing books for their libraries), this is it.
Second, the number 26 seems completely arbitrary to me. I'm not privy to any of the logistical tinkering or number-crunching the folks at HarperCollins did in order to arrive at this figure, but it seems to me (and you can see from the video in the article) that physical books last much longer than 26 check-outs (and I can tell you from experience that they consistently survive many more).
It's possible that this is an average number that takes into account all the books that go on shelves and are never or rarely checked out, left to rot over a fifty-year period with only a couple of loans. If that's the case, though, libraries would chuck the book when it became unusable and would probably not buy a new one, since no one wanted the original to begin with.
If a publisher is going to enforce a loan cap, I think it 1.) needs to be much higher than 26 (based entirely on my uninformed opinion), and 2.) should vary depending on the work in question. Harry Potter and the Seven Figure Advance is going to be borrowed a lot more often than Actuarial Mathematics for Dummies, and the loan cap should reflect this.
Finally, the whole reason the loan cap exists is as an analogue to the wear and tear suffered by physical books that eventually need replacing. I think one of the most destructive tendencies inherent to the publishing industry is not its resistance to electronic media, but its slavish insistence on making them exactly like physical media.
Instead of panicking over a perceived loss in revenues caused by books that no longer need to be replaced, publishers should be touting the non-physical nature of e-books as a tremendous boon: "You'll never need to replace this e-book! That means that rather than buy a new copy of the same book, you can spend the money you'll save on more of our e-books."
E-books don't take up physical shelf space, so the limiting factor that once forced a librarian to choose between replacing a popular title that's worn out and purchasing a different title—that is, space—no longer exists. More titles sold is good for everyone.
My rant is over for today, meine Autoren, but what do you think? Should publishers be able to cap library loans, and if so, is 26 a reasonable number?