Re: Derivative of digital image

Georgia Rothacker <gerrc[_at_]rc.rit.edu> wrote:
> >
> > A young imaging scientist in my company asked, "What is the percentage
> > of a digital image (authored by another and downloaded off the web)
> > that must be changed in order for the work to be considered a new
> > work?" My initial response was that only the author has the right to
> > prepare a derivative. The scientist stated that he is sure there is
> > case law that has established guidelines in this regard. Please advise.

Mark Lemley <mlemley[_at_]mail.law.utexas.edu> wrote:
>
> I do not believe there are any "guidelines" on how much must be
> changed, though I have heard many myths to this effect.

Jamie Powers <jamie[_at_]srgpe.com> wrote:
>
> I too know of no U.S. cases sanctioning any fixed "portion or
> percentage rule" and the genesis of this approach might be found in
> some of the fair use studies and classroom photocopy quidelines.
> Percentages were used in describing many instances of photocopying
> (for example, in the context of how many articles from a journal can
> be copied). If anyone is aware of any "bright line" rules on this
> issue, please advise.

Everyone is right except the original imaging scientist. A derivative work may be created only by the copyright owner. Starting with the original work and "changing" just enough to keep you out of trouble is creating a derivative work, as far as I'm concerned. And there is no bright line rule about how much must be "changed" to make a work noninfringing. A subsequent artist has to create a new, noninfringing work ... he can't start out with an infringing exact copy and then see how many changes he can make so that it ceases to look like an actionable infringement. Jamie is right that some copyright opinions have used percentages in weighing how much of a work was copied, or how much of an accused work was substantially similar to the copyrighted work. But no court to my knowledge has ever set forth a "percentage rule." In fact, people who believe in percentage rules ought to reread Harper and Row v. The Nation, 471 US 539, in which only a few key paragraphs copied from an enormous tome were held by the Supreme Court to have been "insubstantial" in comparison to the original work, but NOT fair use because they had taken "the heart of the book". (The court also noted that the infringing excerpts constituted about 13% of the accused article.)

Bob Cumbow
<cumbr[_at_]perkinscoie.com> Received on Wed Jul 30 1997 - 16:52:37 GMT