IEEE Technology and Society Magazine - March 2020 - 73

systems. In a normal market, copyright users would be
able to shop around for the digital rights management
system that best suited their needs, giving rights management vendors an incentive to constantly improve
their wares. But since copyright owners offered licenses
on a take-it-or-leave-it basis, using a digital rights management system of their choice, little or no shopping
around was possible.
Similar practices prevail amongst web site privacy policies and computer software licensing agreements. It might
be argued that we seem to get by well enough, in that billions of people continue to use the products of companies
like Facebook and Microsoft products day to day in spite
of the controversy about privacy on the former and licensing software from the latter. But nor does that controversy
seem to have made much difference to the underlying
practices - in part because it's not easy to switch one's
social networking service provider or operating system. It's
an open question as to whether users would feel so comfortable if these agreements were enforced by smart contracts in the manner that digital rights management
enforced licensing for books and music.

Countering the Objections
Through the 2000s, computer scientists and legal scholars proposed numerous ways in which the deficiencies of
digital rights management might be countered. As for the
criticisms, many of these turn on particular points of
copyright law, or on the specific ways in which copyrighted works are used in practice. Few of them had very
much impact on real digital rights management systems,
or the laws surrounding them.
From a technical perspective, the major challenge was
to design trusted systems that would permit various sorts
of behavior to which people were accustomed from using
CDs and the like, while preventing indiscriminate file-sharing. This included schemes for interoperability between
the trusted systems of different manufacturers [9],
schemes for limited sharing of digital works (as between
friends but not on file-sharing networks) [10], and
schemes for detecting and responding to misbehavior
rather than preventing it outright [11].
From a legal perspective, this included legislation
designed to compel interoperability between trusted
systems (adopted in France) [12], proposals to implement fair use via an escrow service that allowed trusted
systems to be circumvented under certain conditions
[13], exemptions to laws prohibiting the use of "circumvention devices" [14], proposals that copy-protected
media be clearly labeled as such [15], and proposals
that copy-prevention systems be required to satisfy
minimum requirements for the support of fair use [13].
It's hard to know for sure exactly why these efforts
didn't bring digital rights management that satisfied
MARCH 2020

∕

everyone's desires. Maybe some ideas just weren't very
good; maybe some ideas had potential in their own way
but were introduced too late or weren't sufficient in themselves to solve all of the problems; maybe the ideas were
good but couldn't get the political or market support to
put them into practice; or maybe people just weren't
going to like digital rights management no matter how
well it worked. After all, how many people shop for security systems to prevent them from doing what they want?

What Might Smart Contracts Have to Learn
From Digital Rights Management?
Could it be different for smart contracts? For one, the
context is different: it's conceivable that smart contracts
could succeed in some application or another where
copyright proved an impossible hurdle. Maybe mistakes
made with digital rights management can be avoided.
Or maybe ideas that had not properly matured in the
time of digital rights management have an opportunity
to reach maturity in the time of smart contracts.
Consider application areas. Replacing the entire legal
system with automatically-enforced contracts, as some
of the most enthusiastic proponents of smart contracts
seem to hope for, is a very big ask. Digital rights management researchers couldn't do it even for the subset
of the legal system known as copyright law.
One major reason that automating copyright protection proved so troublesome is that copyright law is
immensely complicated and subtle. Other parts of law,
including contract law, are no less complicated. The
idea of simplifying all of this law so that it can be processed by computer instead of lawyers sounds superficially appealing - but consider what might have been
done with copyright if it were to be "simplified."
The first thing to remove would surely be fair use,
since this is immensely difficult to judge, and little if any
headway was made in automating it. The next might be
the principle known as "first sale" or "exhaustion," which
states that someone who buys a physical embodiment of
a copyrighted work such a book or CD can do whatever
he or she likes with that book or CD, but has no obvious
meaning for a computer file. A simple, clean law that can
be enforced by computer might just say that the copyright owner has the sole discretion to determine who has
access to the work and who doesn't, and under what
conditions. In fact, this is more or less what most
deployed digital rights management systems did.
The point of this thought experiment is that all of
those subtleties and complexities reflect the fact that
the world is a rather subtle and complicated place. Laws
try to balance numerous competing interests and
desires, and disposing of interests and desires because
they seem a bit complicated may quickly lead to laws
that serve only the crudest and most powerful ones.

IEEE TECHNOLOGY AND SOCIETY MAGAZINE

IEEE Technology and Society Magazine - March 2020

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - March 2020