On Jun 28, 2007, Jan Harkes <jaharkes@cs.cmu.edu> wrote:
Yup. But will courts see this as a valid excuse for effectively
denying users the ability to modify the copies of the GPLed programs
they run on tivoized devices, even though the means to accomplish this
are based on blocking the ability to run the modified executable,
which happens to not require permission from the copyright holder in
the US?
Even more so when the modified binary provably works just fine on
similar computers, and fails to work on the tivoizing device just
because mechanisms were introduced to stop the user from modifying the
behavior of the software that runs in the device? I.e., it fails
because you changed it, rather than because you broke it.
Couldn't this be understood by a court as an attempt to exploit a
loophole in the letter of the license, rather than as an intentional
permission to restrict the ability to run covered programs, an act
that the license phrases as "not restricted" and "outside its scope"
just because US copyright law doesn't demand permission from the
copyright holder to run programs, which results in permission to run
being outside the scope of a copyright license?
Anyhow, I didn't mean to restart this debate, I just wanted (as stated
above) to withdraw my earlier tentative agreement that "modification
in place" was something that could be forbidden without infringing
GPL.
Maybe it's fair use, especially for a printed book, of which many
copies presumably exist.
Relevance was only to counter the argument that the right to "modify
in place", as some put it, was not covered by the GPL, "because that's
not how copyright law defines modification." I bought that argument
at first, but then I referred back to US copyright law and not only
failed to find supporting evidence for this argument, but actually
found opposing evidence. So I brought it up to make sure my tentative
agreement wouldn't be used in a court as an indication that the
argument holds water.
Yup, that's the idea.
So, the same standards would apply to the binaries: if they were
inaccessible, copyright holder could claim infringement under the same
grounds, right?
Except, perhaps, for this bit:
If distribution of executable or object code is made by offering
access to copy from a designated place, then offering equivalent
access to copy the source code from the same place counts as
distribution of the source code, even though third parties are not
compelled to copy the source along with the object code.
So, let's narrow the scenario to: tivoized machine downloads binary
from protected site, refrains from downloading sources that it could
download, user can still access and copy the binaries, but can't
obtain the sources because the machine opted not to get them.
Now, the user can't distribute the binaries, because doing so without
being able to get the sources to pass them on would be copyright
infringement. Would a court see this as a restriction on distribution
imposed by the distributor? Or by the copyright holder?
--
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}
-
