[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[dvd-discuss] Some opinions on the appellate court's decision (longish)




If you don't want to read it all, read number (3).

Claus

******************************************************


(1) The court's pre-disposition


Apparently the court follows the arguments and logic of our
side up to the point of "code is speech". Whether the court
just found no way of evading the very dense material
accumulated by the various briefs or truly understands and
blesses our argumentation, this is good.

However, the court does not follow its own good logic to the
final consequence; it takes the first opportunity to bail
out of the stream of rationality by a side door.

The `functional elements' serve to hide the responsibility
of the computer user who initiates the action, and puts some
of the responsibility on the code, ultimately blaming the
bad decisions of a computer user on the code author and
denying constitutional protection to the author's speech.

Generally, the judges are not considering arguments for the
appellant by their merits but by their completeness, and
every argument not covering each and every loophole is
dismissed. The particularly sickening point here is that
some of these arguments could not be made in detail at the
trial because Kaplan wouldn't allow.
- They did not get it that DeCSS is part of the DVD player:
  "... without a DVD player" (page 52).
  They have no clue about the software development process.
  They do not take a program as something open, consisting
  of parts, but as a whole indivisible entity which is somehow
  magically created and from then on serves as a black box.
  Such an attitude is frighteningly Orwellian, it allows for
  all kinds of devices controlling our lives.
- They did not take up on the authorization argument, and
  dismissed it way too easily. They have a very strong
  `property' view of DVD contents being the property of the
  movie industry and not the property of the buyer of the
  DVD, even after the act of sale.
- They totally buy the "DeCSS = loss of sale" argument.
- They are not aware of the possibilities of encapsulating
  arbitrary policies, willfulness, injustice, restriction,
  exploitation, and other stuff into computer programs. The
  public must be able to look under the hood there.

I conclude that the court still does not get the feeling that
there is more at stake than a rogue journalist having fun
helping to crack DVD's. So, we have somehow failed on the
public relations front. The public is absolutely unware of
the issues.


(2) Fair use does not help us


Unfortunately the fair use terms are very vague and very unsuitable
for argument in court. Fair use seems to be dead law, except in some
very well established narrow circumstances, relating to old
technology.

In any way, there is a distinction between first amendment protection
to the speech of the code, and first amendment protection to speaking
materials extracted from the movie as fair use; the one has nothing
to do with the other. We probably did not work on that distinction
well enough to sharpen our arguments.

This makes it very easy for the judges to dismiss fair use: Fair use
could excuse an infringement after the fact, but it does not give
anyone the right to such infringement, let alone the right to getting
the best possible quality of access. (In the eyes of the judges).

The shift to regarding the movie contents as being still a property
of the movie studios once the DVD is sold opens a new dimension of
copyright, one that the public under permanent influence of lots of
`intellectual property holders' has apparently already accepted, but
which has not been part of the original law.

The phrase `secure exclusive rights' can be narrowly constructed as
withholding the rights to create copies or produce revenue from those
from the new owner of the DVD for some limited amount of time. Or,
it can be constructed widely to say that every right not explicitely
sold with the DVD remains with the `IP holder', basically restricting
the right of the consumer to watch it on a pre-approved DVD player and
not much else. The court follows the latter line of thinking.

It might be dangerous for us to insist on the fair use argument too
much; what if the supreme court has also made that move and thinks
that the movie is rightfully the property of its creator and the DVD
is just a medium of transporting its contents in a controlled way to
the customer?


(3) Functionality versus Conduct


The court separates the capability of DeCSS to instruct a computer,
as being nonspeech, from the capability to convey information to
the reader, that being speech.

>From this, the court proceeds to conclude that DeCSS can be banned
for its `functionality', which apparently serves as something similar
to the `conduct' that makes speech sanctionable in general. The DMCA,
then, bans the functionality and incidentally the speech.

I have two objections:

Objection 1:
Under these standards, every book containing instructions to do
this or that is functional; it does not matter whether its
functionality is exerted by means of
  a) a person reading it and doing things
  b) a person reading it and using tools to do things
  c) a person reading it and using a computer to do things
By the same logic, any movie can be described as funcional, because
when you put it into a movie projector it produces effects (sound,
light); it's just one click of the 'on' button of the movie projector
away. There goes speech protection for movies.

The court sees a,b as non-functional since the person is
very visibly involved; it sees c as functional since it does
not understand computers and does not wish to. Thus, it
attributes the functionality and device character to the
code alone, not to code-on-computer.

In fact this is the same flawed logic that allows patents to be
applied to pure software. At the will of those in charge, pure
information suddenly turns into a device.

By not defining a device properly the DMCA reaches broadly into
speech land with its provisions, if the first amendment isn't going
to get it out now, speech land is going to get pretty small.

Objection 2:
The second objection deals with the term `functionality' as some
chameleon that is like conduct but isn't in other places. This
blurs the essence: that the alleged conduct isn't Corley's.

When the court drops the constitutional protection of the speech
because of functionality, it misses the crucial point:
The alleged functionality is not part of the speech of the
defendant. Defendant only posted it, he only did pure speech.
The alleged functionality comes into play when another person
willfully downloads the program, and starts it on his/her
computer. This is where the functionality is located.
This is where it can be punished, since it is conduct.
The only conduct that Corley did was posting it, a purely
speech-related kind of conduct.

By attributing functionality to Corley's posting and then denying
his speech rights to do so, the court made Corley responsible for
another person's willful decision to (a) download and (b) use
the code, as if Corley was responsible for someone elses' actions.

To blur the understanding that the conduct in this case is not
Corley's, the term `functionality' serves very well.
Relying on it, the court introduces a very vague idea that code is
something that acts on its own, like the supernatural `intelligence'
or `energy' or `force' or whatsoever in fantasy movies.
In reality, the functionality of code is invoked by a person
willfully acting (in this case, downloading to the computer,
inserting a DVD, and starting the program), or by a person
instructing or setting up a computer to automatically do those
things. This act represents the conduct part, but the court
frivolously (I think)
   a) makes this an inseparable part of the speech where it isn't
   b) constructs the alleged unlawful behaviour of ANOTHER person
      to be a legal ground for dismissing CORLEY's speech


(4) Code and other speech


On the distinctions between computer code and speech directed to
humans:
One is a clearly a subset of the other, so one could say computer
code is a very particular kind of speech.
However, computer code, to exert its functional component, ALWAYS
requires the deliberate setup of a system by a human which
involves:
   a) bringing the code onto the computer
   b) running the steps involved to convert the computer code
      to an executable piece of memory (compiling, linking,
      relocating)
   c) instructing the computer to run the code
Without those steps, the functionality is not in the code.

All these can of course be combined in a mouseclick, but
there is ultimately always a person responsible who sets up
the system.

The court seems too much in awe of a computer where
everything can ultimately be reduced to a single click or
minor keypress to invoke or not invoke an action; that
clearly blurs its reception of issues and makes a strong
emphasis on the functional aspects of code.

As opposed to other forms of instruction, like recipes and
blueprints, where the functional aspect is not so immediate.

That distinction relies too much on current technology,
however.  There is no barrier that separates
computer-interpretable texts from non-interpretable ones,
and no such barrier can be possibly constructed. With the
right amount of effort, any particular text can be computer
interpreted.

Perhaps it would be good to argue that for many forms of
speech (blueprings, cooking recipes) machines can be
constructed which do the same (shove in normed blueprint,
press button, get result); the only technical aspect here is
that while it's technically possible to do so, it's much
easier technically to create a less capable machine and let
the human express his/her thoughts in a different language
which is more suitable to for automatic interpretation by a
machine.

Ultimately these developments bring the expression of ideas
closer to machines; however, if we deny protection to speech
once it comes close to the interpretation by machines, we
will lose on the speech gradually as machines extend their
capabilities, with the ultimate result that people who do
speech have to be concerned that their speech might not be
computer-interpretable by accident or malicious
construction.

The really BIG issue here is this: human ingenuity has so
far fallen short of making computers intelligent,
i.e. enabling them to interpret human communication in
general. I personally think computers will always fall short
of passing the Turing test. Nevertheless, human ingenuity is
by far good enough to make devices which can interpret any
particular kind of speech. No matter what particular
expression, you can always construct a grammar to cover it,
perhaps with only a few similar expressions. Thus, any
particular kind of speech could be prohibited on the grounds
that a computer can possibly be made to interpret it.

To explain that to a panel of judges is extremely difficult;
a simple explanation will give them the impression that they
are shortchanged by sophistry, and a full training course in
computer languages is not possible.

So, the thing seems to boil down to a chicken and egg
problem; courts are inclined to decide based on what was
there first: if the C interpreter is there first, and the
speech done afterwards (in C), it's "functional" and thereby
can be restricted. If the speech is first, and functionality
created lateron, either not a problem (the cooking recipe)
or deliberate sophistry (when creating a particular
interpreter for a particular piece of speech).

I'm at a loss on how to argue this point, how to show that
once the dam is broken and `functional' speech is not
protected, there is no line to stop.

Ultimately continuing in that direction will drive the
people with independently minded creativity out of coding
and will leave the mindless coding robots in the serve of an
entangled net of big corporations in the field; it's denying
a whole class of researchers access to their fields.


(5) Further proceedings

Two things need to be done very carefully:

First, get out of the head of the judges the current
defendent's apparent bad reputation; they are apparently all
a bit biased. Felten would have achieved that, that's why
the dismissal of Felten is so sickening.

Second, raise the awareness in their conscious AND
sub-conscious minds that once this tiger is let loose,
there's no limit to those engineering solutions that extend
the restriction of speech.  I'm sure they will try to fall
back on an argument where "Let's decide this case against
speech; we will deal with further cases as they come along"
without realizing the fact that there is a continuous
problem domain, where the `functionality' can be transformed
steadily into almost any publication of code, and once
infected by the `functionality' virus, the protection of
speech is gone. Finally someone will come up with a five
point clever criterion which establishes when functional
aspects can be used against speech, only to hide from the
rest of the world the stupidity that underlies the concept
ihn the first place.

Comments?

Claus


-- 
Claus Fischer <claus.fischer@clausfischer.com>
http://www.clausfischer.com/