Skip to Content.
Sympa Menu

cc-licenses - Re: What happens to the GPL in FPGA & VLSI implementations?

cc-licenses AT lists.ibiblio.org

Subject: Development of Creative Commons licenses

List archive

Chronological Thread  
  • From: Rob Myers <robmyers AT mac.com>
  • To: Discussion on the Creative Commons license drafts <cc-licenses AT lists.ibiblio.org>
  • Subject: Re: What happens to the GPL in FPGA & VLSI implementations?
  • Date: Sat, 19 Mar 2005 09:54:56 +0000

IANAL but I am a software developer. ;-)

On 19 Mar 2005, at 01:15, drew Roberts wrote:

Couldn't you build a dedicated peice of hardware that acted like MS Office but
had no software per se?

You can write a piece of *software* that acts like MS Office but has no code in it. The important thing for copyright is whether you are copying the original code, not whether your code is functionally equivalent (that's patents). Reverse engineering and functional equivalency aren't an issue. Copying or deriving are.

When Office is written, it is C++ code stored as human-readable ASCII on a hard disk. It is then translated to assembly code, then that assembly code is translated to object code. That is then linked. We're already three steps away from the original "writing". The linked code is transferred to a removable tape drive, couriered to the CD pressing plant, and scratched onto a sheet of glass as a series of pits. This glass is then pressed against metal foil which is glued between bits of plastic.

Once some poor soul buys the CD and takes it home, the metal and plastic is then spun at high speed under a laser, and this is interpreted as data which is copied onto another hard disk. When a user wishes to run the software, bits of the data on the disk are copied as electrical pulses held in optically etched patterns on silicon.

The original copyright has acted virally through all these derivations and translations. Like fan fiction. If we take a Java program that is produced the same way but compiled from bytecodes to native machine code by a JIT on loading, that (machine translated) work is a derivation (a translation) of the bytecodes, and so still copyrighted.

The medium that the code is recorded in isn't an issue any more than whether you photocopy, screenprint, scan or paint a photograph affects whether you're infringing the original copyright. Nintendo's GameBoy range uses software distributed as ROM cartridges, for example. The same game may be available on an old Nintendo 64 cartridge or in a GameCube CD-Rom (Mario 64 for example). Copying one of these onto audio tape would still be copying. Arranging a series of stones in a Zen garden to encode the data in binary would still be copying. Machine translating it into a real language like Lisp would still be copying.

In this instance the medium is not the message, and whether software is stored temporarily or permanently on silicon isn't an issue. This is the same reason that every router on the net is a high-powered copyright infringement machine, btw. :-)

- Rob.





Archive powered by MHonArc 2.6.24.

Top of Page