Min-Maxing Digital Library Lending

November 20, 2012

I am a gamer. One of the games I spend quite a bit of time playing is World of Warcraft, which might best be described as a really pretty, 3D graphical user interface to a series of amazingly complicated spreadsheets. At the highest level of play, every step is theory-crafted using log files of past performance and simulations of potential performance. Drastic changes are made to eke out 1% more damage or healing. Every time WoW releases new content (really called Raids), there is a race to be the first group in the world to beat the raid. You actually get a special in-game award that shows you were world-first to achieve something like that. So groups go to rather extreme lengths (whole groups of players have been known to change their designated group—at $30 a pop!—just because another group had a very small improvement in damage) to become the world-first for killing a raid boss.

This is what gamers call min-maxing—minimizing the negative traits and maximizing the positives. It is heaven to spreadsheet geeks and the bane of the rest of the players, who just want to have fun. Min-maxing turns games into serious business.

With its new ebook program, Penguin is forcing libraries into a min-max position. The publisher has created a rule set that encourages libraries to enact behaviors that are negative to patrons. Consider the rule set as the rules of a game:

  1. You can only have the widget for one year.
  2. You can loan the widget out as much as you want, but only to one person at a time.
  3. You will pay the same amount for the widget if you loan it 50 times, 10 times, or 0 times.

Given that rule set, the way to “win” at the Penguin game is to cut loan periods. (Patrons be damned, I want to loan this book out as much as possible.) One week loan periods moves us from a potential of 26 loans to 52 possible loans. Heck, why not go down to 5-day loans and boost our potential to 73 loans per year? Wait, we could have 3-day loans and get 121 loans! Hey, I noticed that you weren’t reading that book during the day, so I can loan it to a retired person during the day and a working person in the evening. Hourly loans!

Now maybe it is just that I am a gamer and my mind works this way when presented with a rule set, but I don’t like this in libraries.

This is a drastically different proposition as compared to HarperCollins. By saying you get 26 loans over any time period, HarperCollins doesn’t force the frantic thinking that Penguin does by imposing a set time period. The 26 loans on a HarperCollins book may happen in a 52-week concurrent period, but they may also happen over the next 5 years. It would be interesting, now that we are a few years into the HarperCollins model, to look at actual data on how many ebook licenses have “expired” and been repurchased vs. how many are still operating on the initial 26 loans.

My concern is that Penguin’s rule set might have a very negative impact on collection development. By imposing a one-year kill date, Penguin is cutting off the long tail. And here I am not talking about the extreme fringe of the tail (a place where libraries do still thrive) but rather the still-visible thickness of the long tail just a few short steps in. How many potential loans represent an adequate return on investment for a library to buy a one-year license? 10? 5? 3? 0?

About the only way I can see making this into a positive is to have a strong system of patron driven acquisition running behind the system. You could set up the code to show fake queues for books, and then when you got your acceptable number (5? 10? 15?), you could buy the book and start lending it. If coded properly, patrons would never know about the manipulation; they would just see it as yet another ebook queue where they are number 6 (even though they are really number 1 waiting for 4 more people to jump in) and then gosh, they got the book early for some reason. But then what if the queue never fills up? Did the dog eat the ebook? Oh how this web of lies could grow. . . .

So here is my real problem. Why didn’t Penguin and 3M just go whole hog for the per-loan cost? Why try to dance around the issue with this silly sliding-scale per-loan cost masquerading as a one-year hard limit on a license? We can do math. We can figure out that the one-year kill switch means that, unless we can cram more reads into that single calendar year, we are paying more for each read and thus losing money.

Oh yes . . . money. The real winner here is Penguin. This deal effectively moves the financial onus of the traditional remainders market onto libraries in a contrived digital form. We pay full price for the book whether or not it ever gets checked out. And if it does get checked out, if it shows a glimmer of possible use, then we get to pay for it all over again. And again, and again. But either way, Penguin gets paid.

HarperCollins’s decision to limit checkouts to 26 loans was not popular. It certainly represented the first salvo in a barrage of changes to our idea of library ownership and lending. But really, if we look at the math, it is not a bad deal. It is a different deal than print, and it could certainly use some tweaks (Concurrent loans as an option. Please.) but it doesn’t force a contrived economic model into the library space, or at least not as much as recent developments. More importantly, it doesn’t pit librarians against patrons in a return on investment fight over which books to purchase digitally. Libraries retain the ability to engage in just-in-case collection development to build broad, rich collections with which to foster patron discovery over time.

The hard truth is that we are in a period of extended and fluid negotiations. Though I love to hold up other examples (Baen, K–12 publishers, etc.) as success models, the fact is that the Big Six are a pretty closed and unique system. They aren’t going to be as willing to look outside their closed world when considering potential models. So within the Big Six world, it seems we are much better off holding up a model like HarperCollins’ as a more desirable example than those that have followed. Libraries are going to have to compromise in these negotiations as well. At least the HarperCollins model doesn’t push us into a potential world of min-maxing and lying to patrons about fake ebook queues so we can try and get some sort of return on investment from ebooks with a one-year kill switch on them.