NetNews Usenet Archive 1992 #26/NN_1992_26.iso/spool/sci/crypt/4705

Internet Message Format | 1992-11-12 | 8.2 KB

Path: sparky!uunet!stanford.edu!rutgers!igor.rutgers.edu!zodiac.rutgers.edu!leichter
From: [e-mail]
Newsgroups: sci.crypt
Subject: Re: Limits on the Use of Cryptography?
Message-ID: <[e-mail]>
Date: 13 Nov 92 00:58:27 GMT
References: <[e-mail]> <[e-mail]>
Sender: [e-mail]
Organization: Rutgers University Department of Computer Science
Lines: 153
Nntp-Posting-Host: pisces.rutgers.edu

Terry Ritter asks an intelligent question.  He draws all the same boring,
uninformed answers that make up the vast bulk of the discussion that's been
polluting this group for weeks now.

To pick one particular response in this class (I've removed the name because
I have no bone to pick with this specific author - he's not at all atypical):

|| - Terry Ritter
|  - the response

||    The police bust an alleged child molester, and take possession
||    of his PC.  They believe that the hard drive contains a full
||    database of young kids who have been *or may be* assaulted.
||    That database is enciphered.
||
|| Now, your mission, should you decide to accept it, is to defend
|| cryptography to ordinary voters, congress people and newspaper
|| reporters.  You also need to explain to a relative of one of those
|| kids, someone who doesn't own or work with a computer, why the
|| government should "allow" private cryptography which could hide
|| this sort of information.

|
| The police bust an alleged child molester, and take possession
| of his safe.  They believe that the safe contains a full
| list of young kids who have been *or may be* assaulted.  The
| safe is locked.
|
| Now, your mission, should you decide to accept it, is to defend
| locksmithing to ordinary voters, congress people and newspaper
| reporters.  You also need to explain to a relative of one of those
| kids, someone who doesn't own or work with a safe, why the
| government should "allow" private locks which could hide
| this sort of information.

It is trivial to break into physical safes.  There's not a safe in the
world that will withstand a thermic lance for more than a couple of hours.
Even fairly stupid cryptosystems are much, much more resistant to attack
than the best safes.

Secure physical safes are not an important social issue.  If anyone ever
figures out how to make safes as secure as decent cryptosystems, the same
questions would arise about them.

| The police bust an alleged child molester, and take possession
| of his records.  They believe his records contain a full
| list of young kids who have been *or may be* assaulted.  The
| records are written on paper in a strange language no one has
| seen before.  The alleged child molester claims that it is
| in Pre-Babylonian prune-trading language, and he is the only
| person left alive who knows it.
|
| Now, your mission, should you decide to accept it, is to defend
| historical linguistics to ordinary voters, congress people and newspaper
| reporters.  You also need to explain to a relative of one of those
| kids, someone who doesn't own or work with a computer, why the
| government should "allow" someone to study historical languages which
| could hide

The law doesn't deal with bizarre theoretical possibilities.  "Historical
linguistics" is not a social problem.  Should we write laws forbidding
private ownership of phasers with "kill" settings?  (Stun and "heat rock"
settings would presumably be OK.)

| The police bust an alleged child molester, and attempt to
| take possession of his records, but before they could -- he
| flushed them down the toilet.
|
| Now, your mission, should you decide to accept it, is to defend
| toilets to ordinary voters, congress people and newspaper
| reporters.  You also need to explain to a relative of one of those
| kids, someone who doesn't own or work with a toilet, why the
| government should "allow" private toilets which could hide
| this sort of information.

Another stupid argument.  Everyone knows that there are an infinite number
of ways to destroy evidence.  Guess what:  Destruction of evidence is a
crime.  It makes no difference HOW you destroy the evidence.

BTW, if you knowingly boobytrap your safe so that it destroys its contents
when the police lawfully try to open it, you can probably be charged with
destruction of evidence.  (If you boobytrap your safe in a way that kills
a police officer trying to open it, you will be charged and likely convicted
of at least manslaughter.  Think long and hard about explosives, incendiaries,
and similar nasties.)

| The police bust an alleged child molester, and try to take possession
| of his records.  They believe that he has a full list of young kids
| who have been *or may be* assaulted.  Unfortunately he didn't write
| anything down and maintains the list in his head.
|
| Now, your mission, should you decide to accept it, is to defend
| personal thoughts to ordinary voters, congress people and newspaper
| reporters.  You also need to explain to a relative of one of those
| kids, someone who doesn't own or work with a brain, why the
| government should "allow" people to keep public records.

We have chosen as a society, in the Fifth Amendment, to make this one
tradeoff:  Because of all the possible abuses, we have decided that a
person cannot be compelled to testify against himself.  We recognize that
that may allow some guilty parties to go free, but we are willing to
accept that outcome.

However, we've very carefully circumscribed what is protected.  It's what's
in your head.  Write it down, and the piece of paper can be siezed.  Say
it, and anyone who heard it can be compelled to repeat it.

Most people can't remember very much; they need to write things down.  This
human limitation places a limit on how much we have placed out of the reach of
the legal system - another reason why we are willing to make the trade.   If,
indeed, one cannot be compelled to reveal a key that one has memorized, then
strong cryptography changes the whole game:  From the relatively small amount
of information most individuals can hold entirely in their memories, we are
suddenly in a world where essentially unlimited amounts of information can be
protected.  This changes the basis on which we made the original social
tradeoff - every bit as much as more powerful weapons change the basis of
whatever social tradeoff you believe is implied by the Second Amendment.
I would hope no one out there believes that the Second Amendment guarantees a
right to own nuclear weapons....

| The police bust an alleged child molester, and try to take
| possession of his PC.  They believe that the hard drive contains a full
| database of young kids who have been *or may be* assaulted.  The PC
| has been hidden.
|
| Now, your mission, should you decide to accept it, is to defend
| the right to keep property to ordinary voters, congress people and newspaper
| reporters.  You also need to explain to a relative of one of those
| kids, someone who doesn't own or work with private property, why the
| government should "allow" private property which could be moved without
| government permission and used to hide this sort of information.

You don't need to bring a PC into this.  People have been hiding records for
years.  Sometimes they even get away with it.  How far away can the hard
drive be hidden?  If it's anywhere in your house, it will be found - under
a warrant, the police can literally disassemble your house and everything
in it into pieces too small to hold a drive.  It's your problem to put it all
back together after.  Buried it on your property?  Metal detectors are pretty
good, you know.  If that fails, we'll just start digging.  It's been done;
it'll be done again.

It's all a matter of scale.  Sure, you can hide your PC's disk if you don't
intend ever to use it again - if, for example, you get a warning that you
are about to be arrested.  But it's extremely difficult for you to USE it
while keeping it securely hidden - while it's very easy for you to keep it
encrypted while using it.  The first is not a practical problem; the second
may well turn out to be.

Bizarre hypotheticals are pointless.  If people don't believe something has
a reasonable chance of happening, they could care less what to do about it.
Deal with what people WILL see as reasonably likely.

                            -- Jerry