Re: GPL traitor !

[Hyman Rosen]

David Kastrup wrote:
> It isn't.  The essential copy is the copy in the computer main memory,
> and that is the same whether you link dynamically or statically.

"Essential" copy? What in the world are you talking about? Any
copies of programs or libraries made while executing them do not
infringe copyright, by specific declaration in US law.

Copyright law is about copying. The act of producing a statically
linked program involves creating copies of the libraries and binding
them into a collective work. The distribution of this statically
linked executable involves making further copies of this collective
work. It is this integration of GPLed code into a collective work
that allows the requirement that the GPL must be applied to the whole
work, because creating a collective work requires permission from the
copyright holders of the elements, and such permission is separate
from any other permission granted by the rights holders.

A dynamically linked program does not contain copies of the libraries
it invokes when executed. No permission is required from the copyright
holders of the libraries in order to make and distribute the program
because the program does not contain copies of the libraries. What the
program does when executed is entirely irrelevant to what copyrights
apply to the program.

> Whether you deliver a script which does the static linking, or whether
> you call the dynamic linker makes no difference.

If you deliver a script to statically link a program with libraries
along with the non-library components of the program, then this too
does not require permission from the copyright holders of the library,
so yes, that is the same as dynamic linking.

> > First of all, copyright in a computer program exists when that program
> > is written. The nature of the copyright does not change with the
> > execution of the program; even if the program is never executed at all
> > its copyright is no different than if it is executed a million times a
> > day.
>
> So what?

So it is irrelevant what a program does when executed. When you consider
whether you have the right to copy and distribute a computer program, and
which copyright holders need to be involved in granting permission, you
do not consider what the program does or is intended to do when it executes.
It is a piece of text, and the permissions which apply are those of its
author and the authors of any components which are copied into that text.

> > Second, if execution of the program results in some copying of
> > some protected work, that copying is separate from the copying
> > of the program itself.
>
> The in-memory running copy that actually is the intended (and only
> sensible) use and copy is what counts.

This is false, and obviously so. For example, if someone makes
unpermissioned copies of DVDs, those copies are illegal regardless
of whether they are ever played. Copyright law does not prohibit
the "essential" copy which finally appears in the DVD player, it
simply prohibits making copies without permission of the rights
holders. By your interpretation, bootleggers would have the
affirmative defense that captured copies had never been played!

If I give you the benefit of doubt that you are not trolling, you
at best seem to have a view of copyright law that is entirely at
odds with the actual contents of that law, at least in the US.

> > Not least, that copying is done by the person who executes the
> > program, not by the copyright holder of the program.
>
> Nonsense.  The copying is done by the dynamic library loader
> according to the instructions of the program author.

So what? A copy machine makes copies according to the instructions
of the copier manufacturer, but it is the person using the machine
to make unauthorized copies who is liable for infringement, not
Xerox. Again, you seem to have a very skewed notion of copyright
law.

> >     — Notwithstanding the provisions of section 106, it is not an
> >     infringement for the owner of a copy of a computer program to
> >     make or authorize the making of another copy or adaptation of
> >     that computer program provided:
>
> The owner of the copy.  But the one doing the commands for making
> the copy is the owner of the upstream main program.

No, the one "doing the commands" is the one executing the program.

If I am the author of a computer program which upon invocation copies
the contents of the "c:" drive to a writable DVD, and someone takes a
copy of that program and uses it to copy data from a machine which he
is not authorized to access, that person cannot accuse me of having
done the illegal copying!

It is simply not possible to use copyright law to restrict all actions
that you would like to restrict. No matter how noble you believe the
goald of the FSF and the GPL to be, the GPL is a copyright license and
can only apply when copies of covered works are made.

> But it is not at the choice or the will of the copy's owner that the
> creation of the copy happens, but as an integral part of the upstream
> authors program's intended operation, according to the upstream author's
> instructions.

It is through the action of the person executing the program that the
copies happen.

> If the upstream author sells his program as a working program rather
> than a piece of binary garbage that serves no intent except possibly
> instructional intent for people into reverse engineering, than it is
> clearly _his_ intent and will and command that puts the library into
> use.

The actions taken by a computer program when executed are not relevant
to the permissions required to copy that program. You cannot point to
any copyright law (certainly not in the US) that says otherwise - no
such law exists.