Daniel Bernstein: ten years of qmail security

Lawrence Rosen first says that you cannot release a work into the public domain, and then
seems to contradict himself by saying you can do just that by writing a statement 'I hereby
give it away to anyone who wants it for any purpose whatsoever.'  Surely if you can say that
you can equally well say 'this work can be treated as if it were in the public domain'.  And
if it looks like a duck and quacks like a duck... a work which has no copyright restrictions
(either because they have expired with age, or been explicitly waived by the author) is indeed
in the public domain.

He makes a good point that promises are not enforceable (unlike contracts) and can be
withdrawn.  But if you accept that logic, then no free software licence can be relied on,
since they (almost) all claim to be licences and not contracts; there is no consideration you
pay in return for the right to copy the software.  I think there may be some confusion between
a promise of a gift (which can be withdrawn at any time) and a gift itself (which obviously
cannot; I can't give you a bicycle and then a week later steal it back with impunity).

He says, don't accept gifts of software assuming they are in the public domain.  Of course
not.  You need an explicit statement from the software's author saying that it is his express
wish that the software be treated as public domain.  If you have that, it should be
unambiguous enough even for lawyers to understand.

The FSF in <http://www.fsf.org/licensing/licenses/gpl-faq.html> say that it is possible to
disclaim copyright on a work and so place it in the public domain.  Presumably their legal
counsel has checked that page.  So you must decide which lawyer to believe.  For now I'm going
to side with common sense and assume that if djb or anyone else tells you he has released his
work into the public domain, you can take him at his word.

Unless there are any legal precedents of relevance to the contrary, my own view (IANAL) is
that the idea, that someone can sue someone else who has placed source code into the public
domain for damage caused by it, is FUD without practical foundation. Anyone using public
domain software source code for a potentially damaging purpose could reasonably be expected to
have what it does examined by an expert in order to confirm its suitability before using it
for such an application. You might as well try to sue someone else for a published idea or
research which you misapplied and which went wrong when you did so; this course of action
would also not get the litigant anywhere in the courts.

I can imagine an exception if source code for a trojan was placed in the public domain,
particularly if the source features making this program a trojan were obscure, and good
evidence existed that the author intentionally and/or maliciously included these hidden and
potentially damaging features. But I don't think applying a free software license to such code
would protect the author of it from similar litigation under these circumstances either, as
any disclaimers in this license would be considered moot.

Personally I don't think spreading the FUD: that releasing well-intentioned software into the
public domain can make the author liable - will attract any programmer or decision-maker to
apply free software licenses to code who otherwise wouldn't, though it might turn some people
off free software altogether.