1994-08-15 - Re: Seeking Clipper/Telephone Cost Estimates

Header Data

From: Adam Shostack <adam@bwh.harvard.edu>
To: hanson@hss.caltech.edu (Robin Hanson)
Message Hash: 3c296693b503c97f5f9a4cca743983af9b97d142f749379172a00e808ebb2ce0
Message ID: <199408152141.RAA07858@bwh.harvard.edu>
Reply To: <199408142324.QAA17126@hss.caltech.edu>
UTC Datetime: 1994-08-15 21:43:07 UTC
Raw Date: Mon, 15 Aug 94 14:43:07 PDT

Raw message

From: Adam Shostack <adam@bwh.harvard.edu>
Date: Mon, 15 Aug 94 14:43:07 PDT
To: hanson@hss.caltech.edu (Robin Hanson)
Subject: Re: Seeking Clipper/Telephone Cost Estimates
In-Reply-To: <199408142324.QAA17126@hss.caltech.edu>
Message-ID: <199408152141.RAA07858@bwh.harvard.edu>
MIME-Version: 1.0
Content-Type: text/plain

	At the HOPE (Hackers On Planet Earth) conference, there were a
pair of AT&T 3600c telephone Surety devices on display & demo.  The
executive summary is that they sucked.  The docs do not mention
Clipper at all, but they do have an interesting disclaimer about how
AT&T has no responsibility if the government, or anyone else, taps
your phone.

	The device is $1300.00.  This means someone put up 2600 on a
credit card.  (The conference, organized by 2600 magazine, was much
amused.)  Each unit includes a 4800 baud feature rich modem, a clipper
chip (not marked as such), and some adapters to make it work with
various phone handsets.  The unit plugs in between the handset and the
phone base unit.

	The hackers who bought the things had quite a hard time
getting them to work at all.  There were troubles getting it set up so
that it would attempt to go into secure mode, and trouble getting it
to do so reliably once a pair of phones that worked were found.  AT&T
service blamed the problems on line noise, even though the folks
testing had a CO simulator, and were able to link V.fast modems
through it, and also link through the CO.

	To make the unit go into secure mode, one person pushes a red
button.  The unit sends touchtone 2587 (we wern't sure why; someone
suggested as a means of calibrating.  258 are in the same row on the
phone.)  Then the modems do their thing, making modem noises for about
20 seconds (your time may vary; AT&T manual said 10 seconds.)  Once
connected, the sound is very weak.  We in the conference had trouble
hearing when the earpiece was right next to a microphone.  There was
also a roughly quarter second delay (presumably this is for A/D
conversion + encryption) in talking.  This is a longish delay, roughly
equal to an overseas satellite conversation.

	Lastly, if you send a dtmf down while in secure mode, you
summon the clipper demon, which, we were told, sounds like something
out of the exorcist.  You also drop out of secure mode.  Useful to
know if demoing a clipper box.  :)

	I did not catch the name of the speaker who was doing the
demo.  A post to alt.hope.d would probably find the info.

	There were also two honest to god clipper chips sent by a nice
man at Mykrotronix.  (Thanks to John Droach(?)) One was kept as by the
guy who got them, the other was blown up with a small explosive device
to close the conference with a bang.

	They were quite small; maybe 1 cm square, and .5 cm thick.
Manufactured in the Phillipines, too. :)

	Anyway, thats my brain dump on clipper from HOPE.  There were
a fair number of cypherpunks there; anyone else want to offer
additions or corrections?


| For example, my paper last year included the sentence 
|   The current government contractor claims it will offer the wiretap chips
|   for about $26 each in lots of 10,000 [2], over twice the $10 each a
|   competing private developer claims it would charge [11] for a chip with
|   comparable functionality, minus wiretap support.
| as part of an attempt to estimate the direct costs imposed by the
| "clipper" chip.  I recall seeing that they are now offering these
| wiretap chips for $15 each, but can't seem to find the source for
| that.  I'm also told the clipper chips are big VSLI chips, and too big
| to fit into cellular phones which are the main current potential
| market for encryption chips.  Can anyone offer more technically savvy
| and up to date estimates of any of the added costs such wiretap chips
| impose over other encryption chips?