LWN.net Logo

Examining an attack on the GPL

Examining an attack on the GPL

Posted Nov 27, 2003 6:43 UTC (Thu) by vgough (guest, #2781)
In reply to: Examining an attack on the GPL by ekj
Parent article: Examining an attack on the GPL

How are damages determined?

I've heard people (literally - two anti-GPL people at a coffee shop)
throwing out theories that because people were giving their code away for
free, that there were little or no damages.

But, my choice of releasing code under certain terms (the GPL) should in
no way prejudice my ability to license the very same code for money to
another licensee. So how are damages figured for copyright work which
has not been previously sold or licensed for profit?


(Log in to post comments)

Examining an attack on the GPL

Posted Nov 27, 2003 8:08 UTC (Thu) by gleef (guest, #1004) [Link]

Damages would be hard to determine. If you have an established "dual license" business model (eg. Cygnus, Trolltech), you can make some case at it. You can probably come up with some economic or business value to the code copying, by bringing in expert witnesses. This is further complicated by the fact that you really can't claim significant damages in the US unless you've registered your copyright with the Library of Congress, U.S. Copyright Office, something many Free Software developers don't bother with.

The big stick, from what I've seen in terms of FSF enforcement of the GPL, is the threat of an injunction. If you can show that a person is likely distributing your work in violation of your copyrights, it's fairly easy to get an injunction forcing them to stop distribution until all the issues get ironed out (and permanently, if they can be proved to be distributing it and nothing gets ironed out). Since, in most cases the alleged infringer is a commercial entity who is depending on this product for revenue, such an injunction would be very very costly.

Examining an attack on the GPL

Posted Dec 4, 2003 11:43 UTC (Thu) by Wol (guest, #4433) [Link]

Bearing in mind most Free Software developers are not based in the US, why should they register with the US Copyright Office?

I would EXPECT, as a matter of FAIRNESS, that if I sued an American company in America, that the lack of copyright registration would be irrelevant seeing as (a) I wrote my code in the UK and it's covered by UK copyrights, and (b) the Berne Convention applies.

I don't know exactly how it works, but under the convention I should have the ability to sue anyone, anywhere, for copyright violation and have my Berne rights enforced.

If necessary, I might have to register my copyright with the Copyright Office before suing, but no time limits should be applicable to that registration, seeing as if that American company didn't transact business in the UK I would have no means of redress should I rely on the UK courts.

Cheers,
Wol

Examining an attack on the GPL

Posted Dec 5, 2003 0:28 UTC (Fri) by eread (guest, #1918) [Link]

I have studied Australian Copyright Law (though I wouldn't ever consider myself an expert - IANAL) and from my reading of if someone has copyright with respect to a given work, then Australian laws will uphold that.

In some provisions, the person must be a "qualified person" for the provision to take effect. But in most general provisions, as far as I can see, Australian courts will respect and uphold your copyright in Australia, even if the copyright comes from an overseas country (probably signaturies to the Berne Convention.

I hope this is somewhat accurate.

Examining an attack on the GPL

Posted Dec 5, 2003 0:29 UTC (Fri) by eread (guest, #1918) [Link]

A Qualified person being one who is an Australian, or who created the work in Australia.

I don't think one need be an "expert in the field" like some patent law.

Determining Damages

Posted Dec 4, 2003 17:11 UTC (Thu) by HalfMoon (guest, #3211) [Link]

How are damages determined?

I'll assume you don't mean punitive damages, but just ones based on some rational economic basis.

One ceiling would be all the revenue associated with the product which was doing the copyright infringement. That'd need to be fair math, not the kind RIAA companies use with musicians, so if the product integrates hardware and software, the hardware portion of the revenue WOULD be counted.

Another ceiling would be the engineering effort needed to re-invent, re-implement, re-stabilize the relevant code. Clean-room. That could easily be much more than the revenue from the particular infringing product ...

Whatever number is used would be modified in at least two ways. One would be reducing the number; it may be hard to argue that all the revenue came from unfair competition based on copyright infringement. Another would be increasing it, as in "treble damages" for certain kinds of behavior or as in accounting for some market position acquired by fraud. (OK, that last may expect too much real "justice" out of most of today's "justice systems", sorry!)

IANAL, but I certainly know that these numbers can be arrived at in ways that an unbiased third party would agree reflect the "value" of the infringement to the party doing the infringing.

Remember, if the original author had wanted nothing back, they'd have used the BSD license. But that's not what they wanted ... they chose a license which limited the scope for proprietary value-add. So clearly, every dollar (or euro, etc) associated with that product which wasn't proprietary value-add, or a going-rate service, came from the infringement. There's a damage valuation.

Copyright © 2013, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds