When word of the Lightning Source POD model got around, quite a few entrepreneurs figured there was gold in them thar out-of-copyright books. The problem with the business model is the "quite a few" bit I mentioned above. The only entry barriers are a thousand bucks for big block of ISBN numbers, electronic versions of out-of-copyright texts and a little formatting. What's even worse, from the perspective of the republisher, is that they can't copyright these books constructed from public domain text. So, if a publisher stumbles onto a big hit with an out-of-copyright book from a hundred years ago, there's nothing to stop a dozen other publishers from rushing a version out next week.
Publishers working the free content angle can come at it from multiple directions, provided they have the time and the funding. I've come across at least one publisher with nearly 7,000 large print public domain books listed on Amazon, though I can't say they are selling like hotcakes. But then again, the main market could well be offline, to libraries, senior centers, etc. The big backlist approach has been adopted by so many print-on-demand startups that I can't even estimate the number, though I must have come across at least a dozen at random. One has just under 15,000 out-of-copyright titles listing, and the big surprise is they aren't all Lightning Source. It's possible the publisher is printing the others themselves, since I saw 4 to 6 week ship times on them, or perhaps they are waiting for an order to format them and add them to the live catalog.
The large print angle falls in the category of funny once. Somebody must have thought of it first and can have all the credit, but they can't have a monopoly so it's hard to see the business model. The publishers with the huge backlists of out-of-copyright books may be better heeled and organized than the pikers with just a few hundred titles, and will probably outlast them on the Long Tail. One thing they all seem to have figured out in a hurry is not to spend a lot of money (or time) on cover design. Either you see thousands of identical covers, except for the text, or a clever use of public domain images in a repeatable fashion.
I wouldn't have bothered writing about this phenomena at all except for the Tom Swift titles from various publishers that caught my eye because they were actually selling! This is the original series, from the 1910's and 1920's. I found them all at my grandmother's house when I was a boy, a whole bookcase full of my father's books which she had saved. The titles for Tom Swift back then were thrilling science fiction, like "Tom Swift and His Motor-Boat" or "Tom Swift and His Motor-Cycle" and even "Tom Swift and His Wireless". In addition to Tom Swift, I remember the original Hardy Boys, The River Motor-Boat Boys (or something similar) and the Boy Scouts fighting in World War I. Now that I think of it, it seems to me that there was a lot more killing than in the newer series. They weren't exactly politically correct either.
I still think there's a business model in out-of-copyright books, but I see it in nonfiction. The publishers who have the smarts to choose the forgotten classics of nonfiction for which they have a modern marketing platform and update them for our time would get the benefit of a proven winner with some copyright protection. The publisher could file a copyright for the new editing and commentary and it wouldn't surprise me if simply updating the English usages throughout a book from the 1800's turned out to be sufficiently extensive to provide very good protection for the version. Given the effort and expertise required to do it and the modest size of the market, I doubt there would be instant competition. I think bringing out a series of updated books within a niche could work well, I just have to find the niche and the books!