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

Re: Defining 'preferred form for making modifications'

Anthony DeRobertis <asd@suespammers.org> writes:

> Well, for executable works, we have things like "For an executable
> work, complete source code means all the source code for all modules
> it contains, plus any associated interface definition files, plus the
> scripts used to control compilation and installation of the
> executable." (GPL 3) and "Accompany it with the complete corresponding
> machine-readable source code..." (GPL 3a)

Right, and in the cases we are talking about, the corresponding source
code is going to be more than just what can be automatically

> Also, all of GPL 3 is about the "Program (or a work based on it, under
> Section 2) in object code or executable form." If there is manual
> editing sufficient to create a derivative work (e.g., non-trivial
> edits of a creative nature), it is a "work based on [the program]",
> not a form of the program.

Right, which is why the C source is not sufficient if the object code
came from an edited version of the assembly code.

> Also, please remember arguing against automatic translation is very
> much a two-edged sword: If the corresponding source doesn't have to be
> automatically translatable to the object or executable form, then the
> corresponding source requirement becomes much weaker. Beware creating
> that loophole.

It's got to be the *real* source, not just anything you put down and
say "this is the source".  In the cases we are talking about, most
certainly the modified assembly still has to be completely
distributed.  The question is "does the GPL require distributing the
original C code *also*?" and I see no way that saying "yes" could open a

> No. Not at all! Changing a string constant is nowhere near the kind of
> change I'd demand for considering the object file source. That's
> clearly just an attempt to subvert the license, and I think any
> reasonable person would see through that.

Sure, but I think the same of a few bit tweaks to a gif, or applying a
blur transform or... well, almost anything simple.

So in the original case, my point remains: that the .xcf really is the
source (or part of it), even if a given person happens to want to
tweak bits in the gif--and even if they in fact did tweak some bits.

Reply to: