DRM technology and policy

[John S. Denker]

On 04/22/2003 05:36 PM, bear wrote:
> Material which is coyrighted
> must enter the public domain when copyright expires.

Says who?

That's not the law.
I see no reason to make that the law.

My unpublished notes are protected by copyright.
They are also private.  The copyright will eventually
expire.  It is not guaranteed or even likely that the
material will become public at that time.

(Note that patents are different from copyrights.)

=======================================
Earlier I wrote:

 >> Copyright is about recouping the _fixed_ costs.  This
 >> is necessary in general, and particularly important
 >> when the market is of limited size, so that the fixed
 >> cost per unit is significant in absolute terms.

To which Derek Atkins responded:
> The problem is that the current business models recoup this
> cost by requiring payment on the assumption of a scarce resource.

If we're going to go down that road, we need
to be much more specific about what assumedly-
scarce resource we're talking about.

Suppose I've written a book.  I'll give you a
license to publish it if you pay me $1.00 per
copy.  You can have a thousand-copy license for
$1000.00.  You could get a six-billion-copy
license for even less than $6e9, since I would
give you a quantity discount.  So there is no
shortage of licenses.  Far from it.  AFAIK there
is no shortage of paper or ink for making books.
So I see no reason for believing that real
shortages or 'assumed' shortages play any
significant part in the business model.

So much for books.  It turns out that CDs almost
but not quite fit the same pattern.

As I have mentioned before, the thing that makes
present-day intellectual-property issues tricky
is that they collide with fundamental notions of
privacy.

More than a few people believe that what they
do in the privacy of their homes should be, with
rare exceptions, unregulated by the law.

Until recently, privacy did not conflict with
patent and copyright, because it was uneconomical
to infringe things on a small scale.  Economically
significant infringement required mass production,
which had no pretense of privacy.

The thing that has changed is that now people
can engage in rampant copyright infringement
using the same technology that they use for
private activities.

Business has never assumed a scarcity of printing
presses -- but it has tacitly assumed the absence
of reproduction systems that are (collectively)
economically significant yet (individually) small
enough to hide behind a claim of personal privacy.

There is a profound dilemma here.  I don't claim
to have all the answers.  I'm just trying to point
out that we won't find the answers until we have
a better understanding of what the question is.

Incest is forbidden by law and custom, even
when it involves the same physical activities
that are ordinarily considered strictly private.
The problem is that acts that seem private at
the time can have serious impact on the larger
community.

The analogy is not perfect, but you can see
the point:  Copyright infringement is a bad
thing, and it doesn't become unbad just because
it is technically possible to do it in a sparsely
distributed way so that the individual acts seem
private at the time.

We sometimes disallow seemingly-private acts if
they harm society, even if the harm is distant
in time and space.

=================

We have here a bundle of hard problems.  Some
of them are manifestly intractible, but others
are only moderately hard.  Comparable problems
have been solved in the past.  For example:

The invention of radio overturned previously-
valid assumptions about copyright.  After much
to-ing and fro-ing, the laws were changed to
provide for "mechanical licenses" and for
something that we might recognize as a primitive
micropayments system:  BMI and ASCAP pay artists
based on statistical estimates of how much
airtime each work is getting.

If we solve the new problems right, there is
tremendous upside potential.   Right now
typical web content is produced by amateurs
and hobbyists.  That's fine for some things,
but not for everything.  Imagine how much
niftier the web would be if people who created
good stuff actually got paid to put it on
the web.

Maybe cyberspace is intrinsically barbarous,
such that everything that can be stolen will
be stolen, and everything that remains will
be subject to onerous and intrusive controls.
The real world doesn't work that way, but
maybe cyberspace is intrinsically worse.

OTOH, maybe cyberspace will turn out OK.

We might even be fairly close to a solution.
As an example:

You know how booze bottles and cigarettes
packs come with tax stamps on them.  One
could imagine a simple certificate, possibly
something quite light-weight, certifying that
so-and-so paid the royalty on such-and-such
song.  Then you need a way to buy certificates
at a reasonable price.  Then you need a grace
period for people to get into compliance.
Then you make examples of a few people who
didn't bother to comply.

It's more than an honor system, because it
can be audited.  It's sort of like the
driver's license system.  There is no fancy
DRM that prevents you from driving without
a license, but if you get caught you might
suddenly realize it would have been better
to get the license.

That's just a sketch of an example.  Probably
better schemes exist.  Wild extremes to the
left and right are also possible, but if we
keep our wits about us we should be able to
prevent the wild extremists from taking over.


---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at metzdowd.com