Re: Crypto Mini-FAQ

From: Douglas A. Gwyn (DAGwyn_at_null.net)
Date: 04/30/03 

Date: Wed, 30 Apr 2003 04:25:09 -0400

Roger Schlafly wrote:
> We seem to have a terminology hangup. I'd say that a stream cipher is
> a one-time pad with the key stream simulated by a PRNG.

I still object to identifying stream ciphers with KG systems,
which are only a subset of stream systems. The essential
characteristic of a stream cipher is that as each elementary
unit (character, bit, whatever) is fed into the encryptor or
decryptor, one elementary unit emerges. Block ciphers differ
in that they operate on a substantial collection of units as
a batch.

One-time pad systems employ an amount of key at least as
long as the plaintext, and arrange to avoid re-using the same
key material. They are not necessarily additive (or XOR)
although that is most common, and they do not necessarily
employ truly uniform-randomly generated key material,
although that is most desirable.

>>Quantum cryptography has nothing to do with quantum computing...
> Not much, except that quantum cryptography needs a quantum
> computer in every repeater and router.

No! Quantum cryptography merely exploits quantum coherence
to detect intrusion, or meet some other cryptographic goal.
"Quantum" *equipment* is required, but that equipment does
not by any means qualify as quantum computing. The two
really are quite different.

> I don't see much difference. Cold fusion and quantum computing are both
> just ideas. In both cases, the physics says that it might be possible, but
> no
> one has figured out how to do it. Neither is likely in our lifetimes. If I
> had
> to bet on one or the other, I am not sure which I would pick.

Cold fusion was obviously a bogus notion in the first place,
which was thoroughly appreciated by the physics community
but not by the electrochemists who precipitated the big media
circus, and of course not by the general public, many of
whom continue to be misled to this day. The full story can
be found in Taubes' "Bad Science" (which I have read) and
apparently in Huizengas' "Cold Fusion" (which I haven't read).
There are some fascinating lessons to learn about how greed
warps people's ability to evaluate.

Quantum computing, on the other hand, is not only supported
by theory, but has been reliably demonstrated in the lab
(although only on a small scale at present). There are
well-established quantum-computing procedures for certain
important classes of problem. So while claims of the
imminent demise of all PKCS are wildly exaggerated, certain
kinds of PKCS might indeed be more readily comprmised using
QC than had previously been expected. The practical effect
would most likely be to cause longer keys and/or shift to
different forms of PKCS for which no quick QC procedure is
known.

I wouldn't venture to predict what technological evolution
might occur in such a field "in our lifetimes". There were
nearly no computers of *any* kind when I was born, and I'm
not dead yet.

Relevant Pages

  • Re: Crypto Mini-FAQ
    ... I'd say that a stream cipher is ... a one-time pad with the key stream simulated by a PRNG. ... except that quantum cryptography needs a quantum ... comparing quantum computing to cold fusion will appear as ...
    (sci.crypt)
  • Re: Next generation COBOL?
    ... Quantum, and have devoted several years to gathering an understanding of the ... She does Quantum Computing - with each ... > algorithms which do not deal with QuBits. ... >>> Programming Languages for quantum Computers, however, ...
    (comp.lang.cobol)
  • Re: Where do the words come from?
    ... There are testable differences between classical and quantum computers. ... it's also evidence that we *don't* use quantum computing in our ... used to like before the "brain as quantum thingy" meme took hold. ... both?That's one of the arguments in favour of neural networks (or ...
    (rec.arts.sf.composition)
  • Re: Surrogate factoring approach, analysis
    ... > Quantum computing is so technologically challenging ... Quantum computing at its heart relies on mathematics, ... That's a lie. ... It's like I have the heart of the mathematics necessary for a quantum ...
    (sci.crypt)
  • Re: Surrogate factoring approach, analysis
    ... > Quantum computing is so technologically challenging ... Quantum computing at its heart relies on mathematics, ... That's a lie. ... It's like I have the heart of the mathematics necessary for a quantum ...
    (sci.math)