The Development of the Containerization Industry as a
Close Analogy for PRT Advocatesby Nathan Koren
I'd recommend that everybody who hasn't already done so --
read "The Box" by Marc Levinson.
In many respects, the development of the containerization industry
probably represents the closest historical analogue to the development
of the PRT industry. One key point is that the development of containerization
was hindered by the early proliferation of incompatible proprietary
technologies. While this "Cambrian Explosion" of container system
designs meant that a multiplicity of approaches could be tried, with
the best techniques determined via survival of the fittest, it also
meant that investment in container shipping was far harder to
justify. Each ship, port, and associated trucking fleet had to be
customized to ensure compatibility with any individual system. This
meant that every single given shipping route had to be sufficiently
lucrative and stable -- from one end to the other -- to support the
heavy investment of converting to containers. You couldn't just build
a container port or ship and take it for granted that the various
flows of cargo in the area would find a way to make use of it.
In other words, despite the *technological* efficiencies of
containerization, there was no network effect yet operative in global
transport. Every company and every shipping route was essentially a
customized entity unto itself. As long as this was the case,
container shipping accounted for somewhere around 5% of global
transport. The industry only really took off when the various
suppliers got together and created interoperable, standardized
systems. And yes, this involved companies like Sea-land voluntarily
giving up many of their patents. These were hardly communists -- they
were some of the most rabid capitalists the world has ever seen. Once
the systems had been standardized, one could then build a container
port or container ship with the absolute certainty that it would work
for *any* company's system, with *any* shipping run. This made the
financing of such items *vastly* easier to justify -- you no longer
needed to prove the return on each investment all the way from the
producer to the consumer; rather, you could practically take it for
granted that there was demand somewhere out there in the the network,
which would be able to utilize whatever you built.
I'm obviously simplifying things quite a bit, but hopefully you get
the general idea. Whatever revenue the various companies lost by
giving up their patents, they reclaimed 100x over by benefiting from
the network effect, which had previously been out of their reach. But
selling customers on a novel transport system is quite a difficult undertaking.
Institutional customers will demand a system of standardization that
they are already comfortable with, *especially* in the case of highly
novel technology.
Last modified: June 27, 2007
