1995-12-03 - Re: Info on Netscape’s key escrow position

Header Data

From: Hal <hfinney@shell.portal.com>
To: jsw@netscape.com
Message Hash: b8bdfb529fc57d9c149059b78bbf29dd3538f1014e67cf58f74762f94777582f
Message ID: <199512030408.UAA19830@jobe.shell.portal.com>
Reply To: N/A
UTC Datetime: 1995-12-03 04:22:18 UTC
Raw Date: Sun, 3 Dec 1995 12:22:18 +0800

Raw message

From: Hal <hfinney@shell.portal.com>
Date: Sun, 3 Dec 1995 12:22:18 +0800
To: jsw@netscape.com
Subject: Re:  Info on Netscape's key escrow position
Message-ID: <199512030408.UAA19830@jobe.shell.portal.com>
MIME-Version: 1.0
Content-Type: text/plain


I will join the chorus of criticism:

From: Jeff Weinstein <jsw@netscape.com>, quoting Jim Clark:
> I said that if we are to use this encryption technology in business, we must
> have a better solution than to limit keylength or put keys in escrow.

I don't understand this.  What is the better solution?  No other solution
seems to be discussed by Clark.  Most of his message is devoted to
rationalizing the inevitable changeover to key escrow, which he just
dismissed as unacceptable.

> All
> governments of the world have a valid concern about terrorism and other
> activities of concern to the security of their nations. All of them will
> continue to restrict our ability to provide products to their markets unless
> we build in some mechanism that allows them to legally access
> information that is in the interest of their national security.

This isn't true!  The US government (and I believe most other Western
governments, France excepted) does not presently provide any
restrictions on providing products to US citizens which have strong
cryptography.  There are serious constitutional questions about whether
it could ever do so.  Clark's message seems to be based on the
assumption that legal restrictions on crypto are a fait accompli.
Nothing could be further from the case.  What makes me mad is that his
messages seems to promote an attitude which could increase the
likelihood of these kinds of restrictions.  If people think the battle
is already lost, they will be less likely to fight.  IMO this is going
to be a big, knockdown fight and the eventual outcome is far from
certain.

> A lot of ordinary citizens are rightly concerned about their own privacy. I
> am one of them. I do not want the government to snoop on me, but in fact the
> government, through the FBI, can now tap my phone without my knowing it by
> simply getting sufficient evidence that I am conducting illegal activities,
> then presenting this evidence to a court to get permission. I have no say in
> the matter.

Again Clark is preaching acquiescence.  We have no say in the matter.
Our phones can be tapped any time the FBI wants.  What is the relevance
of this to the issue of network communications privacy?  Doesn't this
again sound like a justification for giving up the battle before it is
joined?  Where is his righteous indignation?  Where is the recognition
that the right to tap communications is not granted by God but an
accident of technology, one which can be taken away by technological
progress as easily as it was granted?

> If we as a company were to take the position that in no case will we allow
> a government to get access to our encrypted messages, or refuse to allow
> key escrow with our products, the governments of the world will quickly put
> us out of business by outlawing the sale of our products in their countries.

False!  I can open a company today in this country and take exactly that
position, and the US government will NOT put me out of business.  What
country is Clark living in?

> The fundamental issue is how do we accommodate the requirements of governments,
> while protecting our rights as citizens.

As I wrote in another context, when a question is framed in terms of
conflicts between the rights of governments and citizens, it is based on
a totally misguided premise.  There are no conflicts between the rights
of governments and citizens in our country.  The only rights are those of
citizens.  The real issue is the conflict between the rights of the
citizens to privacy and freedom versus their right to security and
safety.  I think we all know what Ben Franklin had to say about that.

> None of this represents the position of Netscape with respect to what we
> will do. But if we do not come up with a solution to this problem that is
> acceptable to each government, we will not be able to export our products,
> except with a short key length (e.g. 40 bit keys), and that will not be
> acceptable to corporate customers in other countries. They will create their
> own solution, and we will not be able to sell to a larger world market. In
> fact, we could even be ordered by our own government to establish a key
> escrow system for its use inside the US.

Again Clark attempts to anticipate the advent of a totalitarian style
system of controls on access to cryptography in this country.  Should we
really base our policies on the assumption that this will actually
happen?  Will the American people stand by for such an unprecedented
invasion of privacy?

Some governments are capable of all kinds of evil restrictions on
products.  Is Netscape committed to building in provisions so that their
software won't access sites owned by Jews, so they can sell in Arab
countries?  Obviously they will draw a line somewhere.  I urge them to
consider the moral issues involved in endorsing Big Brother GAK systems
before accepting them as just another cost of doing business.

> I chair an industry group called the "Global Internet Project", with members
> from almost twenty companies, including companies from Asia and Europe. This
> was the central issue we all agreed upon this morning, and we are putting
> together a policy statement whose purpose is to educate lawmakers on the
> importance of quick resolution of this matter.

I am afraid that what the companies really want is global consistency.
That way they can use one set of policies for all countries, and no one
company can get a competitive advantage over others by producing stronger
privacy protections, because they will be forbidden by law to do so.
Whether the policies protect freedom and privacy or not is not really
relevant from this view.

If this is the way things develop, I predict that it will not be
acceptable to the general public.  Netscape more than anyone has seen
how much pressure can be brought through a public perception of weak
software security.  Our own brute force key hacks as well as the RNG
seed problems have well demonstrated that.  Do you think the same thing
won't happen, only far worse, if the government tries to force weak
software down people's throats?

I understand that Jeff has stated that Netscape is actually opposed to
GAK.  It would have been nicer to hear that from Jim Clark, in
unequivocal terms.  The overall tone of his message, as I have pointed
out above, is one of accommodation and compromise with government
restrictions on the rights of free citizens to communicate securely.
He almost seems to think that free strong crypto is already illegal.  I
think he needs to take a good hard look around and remember that he is
still a free citizen of the United States.  My guess is that he has
spent too much time in the company of law enforcement people.  He had
better start trying to understand the grass roots members of his market
if he wants to continue to succeed.

Hal Finney
hfinney@shell.portal.com





Thread