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

Re: RES: What makes software copyrightable anyway?

On 5/21/05, Raul Miller <moth.debian@gmail.com> wrote:
> On 5/21/05, Michael K. Edwards <m.k.edwards@gmail.com> wrote:
> > While these categories are exemplary and can be bent to cover
> > previously uncontemplated forms of work, judges are quite aware that
> > they do so at their peril.  If you think that sister circuits'
> > critiques of Rano v. Sipa Press are brutal, you should see what
> > happens when an appeals court gets something wrong and it has
> > significant public policy impact.  What do you think would be the
> > consequences of holding that a software vendor can win a copyright
> > infringement suit based on an arbitrary combination of their software
> > with other independently developed stuff?
> I think you're talking about something different from what I'm talking
> about.  I'm talking about cases where the program as a whole
> is granted special copyright protection beyond that of its individual
> components.  (That's a literal quote from the text you were responding
> to.)

Special copyright protection on what basis?  I've discussed "mise en
scene", and explained why it doesn't apply to software works that
don't contain a "story".  I've given the basis on which I believe the
use of APIs as APIs renders them uncopyrightable.  I've pointed out
that the Computer Associates v. Altai
abstraction-filtration-comparison test is the law of the land, and
that, as a rule, use of one independently developed component by
another doesn't qualify as "copying" under that test in the absence of
true plagiarism.  Do you have any arguments left that are founded in
law and reason?

If you're just saying "maybe there's a reason no one has thought of",
fine; but that's just another way of saying "I fear the FSF's
unsubstantiated assertion", and I don't particularly see why Debian
should yank Quagga (or any comparable case) on those grounds.  IANADD,

> > The Ninth's use of precedents from the Second, and the citation from
> > Nimmer, suggest that this is a principle of interpretation of
> > copyright licenses that applies US-wide.  I expect that similar rules
> > apply elsewhere in the world (IANALIAJ).  So we can all forget about
> > termination for trivial breach, especially if it's inadvertent and
> > cured at a reasonable stage in legal proceedings.
> Sure.  To understand this issue, it's probably worth looking at
> section 1-106 of the UCC: Basically, in breach, the point is to
> make it as if the breach had not happened.

For those reading along at home, the Uniform Commercial Code is a
"model law", governing primarily the sale of goods but also some other
kinds of "commercial paper" such as letters of credit, adopted to a
great degree by most states in the US.  The table of contents (
http://www.law.cornell.edu/ucc/ucc.table.html ) clearly does not
include anything resembling copyright license agreements.

UCC section 1-106 may be found at
http://www.law.cornell.edu/ucc/1/1-106.html .  It is titled "Remedies
to Be Liberally Administered".  It has nothing to do with standards of
breach at all, and certainly not with regard to copyright licenses,
which are addressed by federal court precedents such as those I cited
from Rano v. Sipa Press.  Indeed, the only kinds of breach said (by
the secondary sources I have skimmed) to be discussed by the UCC (I
have not read its entire text) are breaches of various forms of

In other words, we're haring off into utter irrelevancy again, as we
did with Heritiers Huston.

> In a truly trivial breach (de minimus) court isn't going to care at
> all.

That's not what "de minimis" means.  "De minimis" means that the
amount of copied material is so small that it simply isn't "copying"
-- let alone infringement -- for legal purposes.

> In some breaches, the court can declare that the infringement
> was fair use (or find some other reason that the infringement
> was legal).

Are you totally clueless about the difference between "breach of
contract" and "copyright infringement", or are you deliberately
bullshitting?  "Fair use" is a doctrine applied solely in evaluating
copyright infringement claims and has nothing whatsoever to do with
breach of contract.

> In minor breaches, the court could easily declare that the
> infringement constitutes a grant of license under the terms
> of the GPL by the infringing party .

No, the court could not easily declare anything of the kind.  The
remedies available for copyright infringement are set by 17 USC
chapter 5.  The only means by which a court could encourage -- not
compel -- the creation of a license that had not already been granted
would be to issue a conditional injunction of the style rendered in
Sun v. Microsoft.  In the scenario you offer, that's "distribute with
source code, under GPL terms, or don't distribute at all".  It might
be possible to compel specific performance in an action for breach of
contract or under some state tort law such as "unfair competition",
but not AFAICT under copyright law.

> The GPL's termination clause only would kick in where none of
> the above could hold -- and I think we can agree that that would
> not be a trivial breach.

You're standing the law on its head.  Termination of a copyright
license for breach, per Rano and op. cit., can only be done if the
breach is "of so material and substantial a nature
that [it] affect[s] the very essence of the contract and serve[s] to
defeat the object of the parties".  It is inconceivable to me that
I_WANT_OPENSSL could meet this standard.

> > As for termination at will -- I doubt that any court in any
> > jurisdiction would permit such a thing with respect to the GPL, but
> > there is certainly room for a tactical choice of law and venue there.
> > That's why sane people write "choice of law" provisions (not
> > necessarily "choice of venue", which is controversial, and often
> > ignored in "forum non conveniens" proceedings) into their licensing
> > agreements, not to mention explicit "term and termination" clauses.
> > The GPL, again pretending not to be a creature of contract law, omits
> > them, to the great annoyance of people who want to know the rules of
> > the game the FSF is playing.  But the FSF likes it that way -- FUD is
> > their stock in trade.
> You've been fairly free with your criticisms of the FSF.
> Usually, this has been in the context of an obscure legal claim which
> on close analysis seems to be saying something at odds with your
> criticisms.
> Personally, I find this annoying.

There's nothing obscure about this claim, my friend.  And I doubt "FUD
is the FSF's stock in trade" is half as annoying to you as your
periodic snatching at irrelevancies is to me.

> > > And I think we can agree that, at least within the U.S., this definition
> > > is a part of copyright law.
> > >
> > > [On the flip side, if it can be shown in court that there's some criteria under
> > > which all programs are free of copyright law, that's probably a good thing
> > > for the free software community.]
> >
> > Have you given more than a moment's thought as to what would come of
> > that, in the absence of a new and better software rights law to
> > replace it?  For starters, bye-bye GPL, and bye-bye all other open
> > source licenses -- you publish it, it's public domain.  Next, bye-bye
> > software industry as we know it; and while you might think you would
> > like that, you may think again when your telephone network and your
> > electrical grid and your banking system all collapse because half of
> > the vendors of their ops software can't make the transition to your
> > Brave New World.  I'm not going to bother arguing the rest of the way
> > to the Death Of Debian (TM).  :-)
> Do you not classify this as FUD?

What do you think would happen if software copyright vanished
overnight?  I classify this as "reductio ad absurdum".

> If a court finds that there is some context where copyright does not apply
> to any programs, the scope of that precedent would certainly be far
> narrower than what you've suggested in this paragraph.

Did I misread this:

> > > [On the flip side, if it can be shown in court that there's some criteria under
> > > which all programs are free of copyright law, that's probably a good thing
> > > for the free software community.]

I can't find any way to read that other than a fantasy of a
software-copyright-free world, but perhaps you had something else in

- Michael

Reply to: