One of the major obstacles to innovation is switching costs. Here’s a story that shows why: after 120 years, the main library at Princeton University is finally converting all of it’s books to the Library of Congress book classification system.

This is remarkable for several reasons.

The main one is that the Library of Congress system itself is 113 years old!

Princeton’s system had been invented about 7 years earlier by University librarian Ernest Cushing Richardson. At the time, academic libraries had decided that the Dewey Decimal System, invented in 1876, was too simple for research libraries. Harvard, Yale and Princeton each developed their own (and I’m sure that today, each of them will tell you that theirs was first…).

Richardson developed his system in the early 1890s, and all of Princeton’s books were duly filed according to it.

The Library of Congress system was designed for research libraries was introduced in 1897, but then it was too late for Richardson. At the time, he said that he thought that eventually all libraries would use it. But he wouldn’t, because switching all of the books over would be expensive and time consuming. And, he’d just reclassified all of the books once to reflect his new system.

Princeton didn’t start using the LoC System until the late 1960s (around the same time that they finally got around to admitting women). So for the past 40+ the library has held books using both classification systems. This has led to books on the same topic being filed in completely different physical locations – a problem. This is what motivated the final conversion over to LoC.

It’s an innovation diffusion story that has played out over 100 years – which illustrates a major problem in diffusing innovations – it is often costly to switch. Richardson acknowledged that the LoC System was better. It was simply to expensive to change to it.

It’s not enough to just come up with a great new idea, even one that is clearly better than whatever it replaces. You also have to get the idea to spread. Switching costs are one of the big obstacles to doing so.

And what about Richardson’s innovation? It’s not completely dead. Princeton has decided that it’s still too much trouble to convert all of books outside of the main library. So there will still be a few books around using his very long call numbers…