A CC of my letter to Gelernter@cs.yale.edu

I caught your piece on the Op-Ed page. Of course I agree with you about
the fact that there is a need for wiretaps. But I think that there are
serious problems with the Clipper chip. It is very expensive compared
to software only solutions and it is also very, very brittle. 

First the cost:

        * Every extra chip adds to cost, battery usage and pocket bulge.
        This is a heavy price to pay for something that could be done
        in software. Many of the next generation digital phones, for instance,
        use a DSP. There are more than enough spare cycles available to 
        do encryption.

        * The high cost is effectively a tax on privacy for the average
        person. It is easy to imagine clipper chips adding $100 to the
        cost of an already digital phone, fax machine or computer. The
        poor are just as much targets as the rich. In fact, they are often
        easier targets because no one wants to bother chasing down a
        fake credit card transaction or bank account withdrawl if it
        only amounts to $75. If a chip was necessary, then it would 
        be a different story. Software could help all of America,
        not just the rich.


Now the Fragility:

        * The first generation of Capstone chips is already obsolete.
        They're worthless. All of the money that went into fabbing
        and producing them is gone. Why? Because the NSA discovered
        a weakness in the Secure Hash ALgorithm. They've fixed it
        now, but all the old chips are worthless.

        * Imagine that problems arise well after the chip is standardized.
        What will millions of Americans do? All of the digital phones,
        fax machines and modem cards will need to be replaced.

        * Now imagine that a pair of turncoats sell out America and
        put both halves of the key escrow on the black market. If we're
        lucky enough to discover this leak, it could easily take 6 months
        to a year to replace our now worthless phone system. 

        * Software, on the other hand, is very easy to change. In many
        cases, the anti-virus programs travel faster than the viruses.

These are the main reasons why I think that the Clipper is a boondoggle.
Software based solutions would solve all of these problems _except_ the
government's desire for a firm grip on the world and technology. 

-Peter Wayner

Peter Wayner writes:
 >         * Imagine that problems arise well after the chip is standardized.
 >         What will millions of Americans do? All of the digital phones,
 >         fax machines and modem cards will need to be replaced.

Not that I don't agree with the basic premise, of course, but there's
a similar risk to *any* consumer electronics implementation of a
cryptosystem.  Of course, in Clipper/Skipjack's case the problem is
magnified by the fact that the stuff is kept secret, but the potential
remains. 

 >         * Software, on the other hand, is very easy to change. In many
 >         cases, the anti-virus programs travel faster than the viruses.

However, a software-based consumer communicator will probably end up
implying at least as much weight in people's pockets, and as much
extra money, as Clipper.

I don't think an economic argument will really fly well, though I'd
love to be shown to be way wrong.  Seems to me that a mass-produced
chip whose production is subsidized by the government would probably
be pretty cheap.

--
| GOOD TIME FOR MOVIE - GOING ||| Mike McNally <m5@tivoli.com>       |
| TAKE TWA TO CAIRO.          ||| Tivoli Systems, Austin, TX:        |
|     (actual fortune cookie) ||| "Like A Little Bit of Semi-Heaven" |

Mike McNally says:
 > love to be shown to be way wrong.  Seems to me that a mass-produced
 > chip whose production is subsidized by the government would probably
 > be pretty cheap.         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
   ^^^^^^^^^^^^^^^

Your tax dollars at work.

     Dave Otto -- dave@gershwin.jta.edd.ca.gov -- daveotto@acm.org
"Pay no attention to the man behind the curtain!"         [the Great Oz]
	    GAT d++(-)@ -p+(---) c+++ l u++(-) e++/* m++(*) s-/+ !n 
		    h---(*) f+ g+++ w+ t++ r+ y++(*)

Dave Otto (the Wizard of TOTOSoft) writes:
 > Mike McNally says:
 >  > a chip whose production is subsidized by the government would probably
 >  > be pretty cheap.         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 >    ^^^^^^^^^^^^^^^
 > 
 > Your tax dollars at work.

True, of course, but looking at things as they are today a lot of my
tax dollars are long gone into the pockets of the contractors and NSA
people who did the design.  A commercial product not using Clipper is
at a disadvantage from the get-go.  Seems to me this is already true,
and short of exploiting the "investment gap" created in the Clipper
project by the inherent waste and inefficiency of Government-driven
projects, it'd be tough.  Certainly, there's a tremendous marketing
problem (which, I realize, our evangelism might help ameliorate).

--
| GOOD TIME FOR MOVIE - GOING ||| Mike McNally <m5@tivoli.com>       |
| TAKE TWA TO CAIRO.          ||| Tivoli Systems, Austin, TX:        |
|     (actual fortune cookie) ||| "Like A Little Bit of Semi-Heaven" |