Alexandre Oliva wrote:
On Jun 10, 2008, Les Mikesell <lesmikesell(a)gmail.com> wrote:
> You aggregate something for distribution. It's not a whole unless the
> components are combined. And while this may be a fuzzy area for
> things covered by the stock GPL, the version that covers Linux
> specifically says
Parse error, end of sentence not detected. Please fix and recompile
Sorry, after I looked it up I decided that the word 'user' in the
exception would have been confusing here and left it out, forgetting
about the dangling part.
> No, what exempts it is the fact that they are separate things
> their origin and destination.
And how about during the split-second in which distribution occurs?
Aren't the components combined at that time?
> No, you can take separate works and put them together as long as
> remain separate works - as the stuff loaded into a device's firmware
> is separate from the kernel.
You're looking at only one side of the question.
And you are only looking at the middle when it is aggregated, not when
it originates or is deployed to its separate destination.
The other is, is the
kernel separate from the device's firmware?
That's a different question, unrelated to how that firmware got into the
device. You might make a plausible case that the GPL restricts code
from being run on hardware with non-GPL bios and other firmware. I don't
think I'd agree but it would be fun trying to work out the difference
between calling something in ROM and a linked library.
As of today, it very
I don't think that's clear at all, but I think in terms of interfaces
and that pretty much by definition, one piece of code ends where the
other interface starts - and since using those interfaces is the whole
point of having the copy of code I'd have to consider it fair use of
Or, what evidence can you provide that the kernel is
an independent work from the firmwares, against the various pieces of
evidence that it is dependent on them?
It's an interesting question, but it doesn't really relate to copyrights
or derived works except in the upside down world of the GPL restrictions
where no legal rulings exist.
> Please... try to imagine the time when that was written and
> about tapes full of commercial software instead.
Or rather think about the tapes full of GNU software under GPL, LGPL
and a bunch of other Free Software packages under various other
licenses, some GPL-compatible, some not, built for various operating
I can't think about that without remembering the time when GPL software
could not have existed without a commercial OS hosting it, and that
history makes me believe that the GPL itself cannot prohibit this
co-existence even if it is somewhat less necessary now. Saying it does
would seem like going back in time and killing your grandparents so you
can't exist now...
> In the past the FSF has claimed that something should be
> derived work and covered by the GPL if it needed a GPL'd library to
> function, even if it was not distributed together.
IIRC the reasoning goes like, when it is linked with the library, or
gets code from library header files, the copied portions provide
strong indication that the program was developed as a work based on
the library. And then, the copyrightable copied portions actually
make the resulting work actually derived from the library, even if its
sources and object files weren't. And then, if you use dynamic
linking, this copies far less from the library, but it doesn't change
the status in any meaningful way.
Using header files as needed to make something work was pretty much
established as fair use in AT&T's failed suit against BSDI years ago. I
don't think there was an actual ruling but they gave up on any hope of
winning. The FSF claimed their argument about functionality applied
even if no work was copied.
>> But I think we can agree that _until_ there's a ruling,
>> firmware in the kernel is just a gratuitous risk.
> The same risk goes with any GPL covered work.
Nope. The risk that there might be some unknown restricted portion of
code hiding in a GPLed program is quite different from that of a known
We know the GPL imposes restrictions. That's a given, not a risk. But
it doesn't restrict against aggregating other things.