Re: IBM doesn't like the GPL

[Hyman Rosen]

Alexander Terekhov wrote:
> Think of the following
>    GPL.c (GPL'd computer program work in source code form)
>    NonGPL.c (NonGPL'd computer program work in source code form)
>    aggregated into one file (tarball, ISO image, or whatnot)

GPL.c is being copied and distributed as a verbatim copy.
Nothing further is required.

>    GPL.o (GPL'd computer program work in object code form)
>    NonGPL.o (NonGPL'd computer program work in object code form)
>    aggregated into one file (tarball, ISO image, or whatnot)

The source code for the GPLed portion must be made properly
available. NonGPL.o is irrelevant, because it is combined in
a "mere aggregation". (A rights holder might choose to argue
otherwise, though.)

>    GPL.o (GPL'd computer program work in object code form)
>    NonGPL.o (NonGPL'd computer program work in object code form)
>    aggregated into one file executable.

The source code for both GPL.o and NonGPL.o must be made properly
available under the GPL, because the two are now part of a combined
work, and permission to do this is given only if the work as a whole
is distributed under the GPL.

What's your point? Do you believe that a judge would have any
trouble understanding these distinctions? Judges aren't generally
stupid, and sophistry doesn't get claimants very far.