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

Re: GPL on rendered images

On Mon, Dec 13, 2004 at 09:52:13PM -0500, Joe Moore wrote:
> As I understand it, programmatically converting the Pascal code to C does 
> not introduce any creative element. So as far as copyright is concerned, 
> the C code is the exact same work as the Pascal code.  (Just as the object 
> or executable code is the same Original Work)

This means that I don't get a copyright claim in the document merely by
converting it, yes, but that's irrelevant to whether it's the preferred
form for modification.

For example, running source code through an obfuscator is also a non-
creative mechanical process, just like converting pascal to C, but the
result of that is probably never going to be source code.

On the other hand, running the code through a pretty-printer, which changes
formatting but not the language, gives output which can reasonably be
used as source, discarding the original.

> In order to distribute the derivative work, you must distribute under the 
> terms of the GPL.  That means that you must include (or offer) the 
> "preferred form for modification".
> The preferred form for the Original work is Pascal.  The preferred form for 
> the new (combined/derived) work is C.  I think you would need to distribute 
> both to comply with the GPL.
> Another way to look at it is:
> The Original Work has its "preferred form" set in the GPL at license time 
> by the Original Author as Pascal code.  Since you can't change the license, 
> you can't distribute the Original Work (as embodied in parts of derivative 
> works) without that form.

I disagree.  Nothing in the GPL says that the preferred form for modification
is fixed to the language originally used by the author; it says "preferred
form for modification", not "preferred form of the copyright holder for

FWIW, a previous incarnation of this discussion:


which I don't have time at the moment to review.  (Hmm.  I wish the list
archives gave a link for each entire thread, containing just that thread
in mbox format; it would make thread review much quicker ...)

Glenn Maynard

Reply to: