Cryptography Reference
In-Depth Information
In a similar vein to Simmons's “information integrity” program, Gold-
reich's quest for firm foundations appears a primarily conservative endeavor.
It proceeds from a definition of cryptography committed to the prevention
of cheating: “Cryptography is concerned with the construction of schemes
which are robust against malicious attempts to make these schemes deviate
from their prescribed functionality.” These prescribed functionalities would
originate from cryptographers' “definitional activity,” which Goldreich
describes as “the identification, conceptualization and rigorous definition
of cryptographic tasks which capture natural security concerns.”
46
The
simple and elegant lines of such a characterization stand in contrast with
the inherent complexity of Simmons's project: could the myriad issues
inevitable in the design of digital analogues to paper-based protocols—
from signatures to voting and cash—be reduced to the capture and iden-
tification of “natural security concerns”?
In any case, cryptographers heeded the call and embarked on the
arduous task of providing clear definitions of the entire cryptographic
toolbox. The toolbox included simple primitives (e.g., block ciphers, hash
functions), the building blocks from which more complex systems could
be built, but also protocols, whereby several participants interact in order
to realize a security objective, for example, the creation and verification of
a signature. Although the task of proving the security of basic primitives
has met with some levels of success, proving the security of protocols has
proved more challenging, a process I discuss at more length in chapter 7.
Notably, none of the major reference works and textbooks published
during this period—that is, Doug Stinson's
Cryptography: Theory and Prac-
tice
, Ron Rivest's “Cryptography” entry in the
Handbook of Theoretical
Computer Science
, and the
Handbook of Applied Cryptography
—include steg-
anography as part of these foundational efforts.
47
In spite of these disci-
plinary commitments, issues relative to the concealment of information
would make a surprising reappearance, in fact, within the very heart of
public key cryptosystems.
Subliminal Channels
If cryptography is indeed about communication in the presence of adver-
saries, then it also about the perhaps less intuitive cases in which the
communicating parties
themselves
are the adversaries. In 1981, Manuel
Blum proposed a protocol enabling cryptography's most famous couple,
Alice and Bob, to flip a coin over the phone (“They have just divorced, live
