Subject: Re: Thought crimes
From: "Stephen J. Turnbull" <turnbull@sk.tsukuba.ac.jp>
Date: Thu, 28 Dec 2000 20:10:00 +0900

>>>>> "kms" == Karsten M Self <kmself@ix.netcom.com> writes:

    kms> I'm not speaking of segments of society, but of agents, power
    kms> centers, and points of control in the publishing market.

Of course I know that.  What I'm pointing out is that whether you
speak of them as such or not, they _are_ part of society.  "I saw you
palm that card."

If you want to argue power of one small segment of society against
another, much larger one, be my guest.  But to set up publishers as a
Dark Force opposed to the whole of society, well, I expected better of
you.

    >> Publishers, as far as I have heard, are not to be granted any
    >> new rights.

    kms> Flatly wrong.  First, there's Lessig's bovinity argument.

Sheesh.  The very name shows you're clutching at straws.  The correct
reading of the proverb is "locks are for honest people", not "locks
are for semi-sentient beings like cows."  But that prejudices the
argument in favor of copyright owners, so Lessig had to rename it.

"Bovinity" has nothing to do with publisher's _rights_.

In context of my argument, those _powers_ (not rights) conferred by
consumer honesty in the face of transaction costs to dishonesty :-)
are still acquired from authors (with the transfer of the copyright).
Publishers are not granted them directly.  If those powers are
effective, then publishers will have to pay authors more for them.
(Subject to monopsony power; talk to the FTC about that.)

The only case where this could possibly amount to a direct grant of
new powers to _publishers_ is in the case of distribution of
previously acquired works on the new copy-protected media.  It's not
clear how important this is; even if you can't use the copy-protected
DVD version of "Green Mile" in your review of Stephen King movies, it
should not be great trouble to acquire a video copy.  But I'll grant
it in theory.

Furthermore, I'll grant that if 17 USC 1201 is effective against
circumvention and/or there exist technical means invulnerable to
circumvention, it amounts to a grant of a new right directly to
publishers.  But once hedged as above, is that so fearsome?

True, there is a lot of power there, according to the bovinity
argument; but eventually it results in incentive for production and
distribution of information, just as the direct enforcement of
copyright does.  The real problem is the accumulation of capital and
monopoly/monopsony power in the publishers, which results in weakened
incentive to produce, and even weaker incentive to distribute,
information.  The danger in weakening copyright to "free" the
information is that although in the short run it frees existing
information, in the long run it may simply reduce the amount of
information produced as long as the monopoly power is not directly
attacked.  It may not increase the amount of "free" information at
all!

(Reducing monopoly power while maintaining copyright is not an
oxymoron.  Of course copyright involves a franchise on a particular
work.  The monopoly power I suggest reducing is not this franchise,
but rather the power that results from a single publisher cornering
the market for a broad set of similar works.  If publishers of similar
works compete against each other, this will result in higher
compensation to authors, lower costs to consumers, and more
dissemination of information.)

Granted, there are going to be "unique" works (eg, the Westlaw (?)
legal reference numbering system).  But in many cases the uniqueness
is created by bad laws or regulations that should themselves be
amended, not ameliorated by messing with the copyright law.

    kms> Second, technical means apply equally to both existing and
    kms> new works -- by releasing works for which rights already
    kms> exist on copy-protected media, the publisher has gained, by
    kms> technical means, advantages protected by law.  The right
    kms> gained is to bypass the 107 Fair Use exemptions on exclusive
    kms> rights.

Again, that is not a "right", that is a _power_.  The point being that
if the technical means are circumvented*, the Fair Use exemptions come
into play.  It is not clear to me that the purposes of the Fair Use
exemptions require a perfect bit-for-bit copy of any portion of the
work.  Nor is it clear to me that capturing the analog output of a
copy-protected digital audio MD, which is a circumvention* for my
purposes, is a violation of the anti-circumvention provisions of 17
USC 1201.  You need one or the other of those before I'll concede that
publishers have acquired "right to bypass".

[* = an economic definition of circumvention which doesn't require a
perfect copy, but merely a sufficiently good likeness that it would
constitute a copyright violation in the absence of "fair use".]

I'm also a little leery of the phrase "right to bypass the
exemptions."  I don't understand "fair use" well enough to contest it,
but it clearly prejudges the issue, in the same way that use of
"honest people" vs "cows" prejudices the bovinity argument.  You
interpret this as acquisition of a new right to override a consumer's
right to fair use.  But it could just as easily be interpreted as
establishing a new context for the same old copyright, where the
consumer happens not to have the power to acquire a copy.  Just as he
hasn't the power to acquire a copy of a movie in a theater where
videotaping the movie is prohibited, although he does have the power
if he acquires the movie on video.  Nothing "new" about the right
itself.  This is presumably what the authors of the White Paper had in
mind when they wrote that there is no affirmative right to fair use.

    >> BTW, are there commonly occurring, plausible examples of "fair
    >> use" of works that can be stored on non-self-revealing media?
    >> Otherwise, the possibility of invulnerable technical means
    >> seems a non-issue.

    kms> I don't understand the question.

This is equivalent to an application requiring a perfect bit-for-bit
copy, a case where the anti-circumvention provisions of 17 USC 1201
by themselves pretty clearly inhibit "fair use."

    >> The fact that authors are in a relatively weak[2] negotiating
    >> position reflects that fact that making gobs of money from a
    >> publication requires massive investment in production,
    >> distribution, marketing, and, of course, copyright enforcement.

    kms> As you pointed out in your Samizdat argument, individuals can
    kms> publish.  The web is one mightily effective publishing and
    kms> distribution tool, and for the time being, it is relatively
    kms> accessible to the lowly individual.  I've met several people
    kms> who've started relatively successful worldwide publishing
    kms> enterprises with literally little more than a few thousand
    kms> dollars capital and a bedroom -- The Register, Slashdot,
    kms> Kuro5hin.

Uh-huh.  Me, I think that's good enough that we needn't worry about
the oppressed consumers.  But I was under the impression you want
something that can look Ted Turner in the eye.  The content of The
Register, Slashdot, and Kuro5hin put together and multiplied by 100
won't attract a takeover bid from Time/Warner.

    >> I note that the White Paper's logic is quite consistent with
    >> the logic by which Stallman consistently refuses to revise the
    >> GNU GPL to prohibit use of modified versions of GNU-licensed
    >> software without redistribution by (eg) ASPs.  Many an ASP is
    >> nothing but a way of preventing the user from making "fair use"
    >> (among other less fair uses) of software.

    kms> Parsing error.  RMS won't prevent ASPs from running modified
    kms> (but undistributed) derivatives of GPLd works?  That's my
    kms> highest-likelihood interpretation of a 43-word sentence
    kms> containing a triple-negative.

Sorry.  That's the intended interpretation.

    kms> If this is your interpretation, there are extralegal means of
    kms> persuading an ASP or other "public performance" user of
    kms> derived GPL works that participating in the free software
    kms> process is a Good Thing [tm].

But we're talking about using the law.  There are extra-legal ways of
persuading Disney, too; that doesn't make the cases different.
Stallman won't write it into a private contract; I think that's a
strong argument against writing into law.

    >> Stallman certainly can't be accused of "being unconcerned", but
    >> he comes to the same conclusion as the White Paper: the
    >> technical means have to be admitted.

    kms> Stallman's reasons fall into the other camp, however.  He is
    kms> concerned with protecting the public's right to access to
    kms> sources, rather than of publisher's right to restrict fair
    kms> use.

No, the reasons are the same.  True, Stallman created the GPL to
protect the right to access sources.  He would dearly like to reach
into the ASPs and force publication of those modified sources.
However, he acknowledges that he can't justify that.

    kms> White hat, in my book.  Robin Hood, not the Sheriff of
    kms> Nottingham.

He dislikes the conclusion as much as the publishers like it.  It is
that very fact that makes Stallman's acceptance of similar logic so
significant.  "Robin Hood bows to the moral correctness of the
Sheriff, in this one case."

    >> I certainly hope that the law will be amended to broaden rights
    >> to fair use as necessary, and that it will not be amended to
    >> make circumvention, as such, of "technical means" illegal.

    kms> In this last matter, you're no doubt aware that it already
    kms> has been?  17 USC 1201 specifically addresses "Circumvention
    kms> of copyright protection systems".  You've heard of Jon
    kms> Johansen and DeCSS?

(1) I was at one time and am again now.  (2) Yes.  I, er, forgot.
Wishful thinking.  Temporary insanity.  Mea maxima culpa.


-- 
University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences       Tel/fax: +81 (298) 53-5091
_________________  _________________  _________________  _________________
What are those straight lines for?  "XEmacs rules."