Metaprogramming and Free Availability of Sources

[Francois-Rene Rideau]

>>: Laurent Martelli
>: Richard Stallman

>> In the absence of intellectual property, people could still distribute
>> binaries without the sources.
>
> Yes, you are right.  Therefore, simply eliminating copyright would not
> make all software free.  To do that, we would also need a consumer
> protection law requiring publishers of software to make source code
> available.

I'd like to differ slightly.
I think there is an absolute natural right to freely use, copy, and modify
any information that one has for any of one's legitimate purposes
(if used with an illegitimate purpose, the attempt to fulfill this purpose,
and not the means of using information, should be punished).
This right is of the universal kind that constitutions should recognize.

But there is no absolute natural right to access other people's information.
And for good. Some information is better kept secret,
like cryptographic keys or randomness pools,
source of programs competing to a contest
(that is, at least until contest application is closed),
or perhaps information coveted by your enemies
(although it is better to not have enemies).
So that I would oppose any law that would demand all information
to be revealed without discrimination.
Secrecy is also useful to protect intellectual workers
against who'd just steal their ideas.

Patents and copyrights were allegedly invented to encourage people
publish their work instead of keeping them secret,
even though we now know that a free market allowing private consortiums
to buy secrets from authors at market value is a much more effective way
to encourage creation of valuable information.
The ability to keep things secrets is nonetheless
as fundamental a right of human beings
as that of using, modifying and republishing information they have.
Recognition of this fundamental ability precedes
the very notion of encouraging publishing of information,
be it by means of either protection or liberty.

I'm convinced that making NDAs (Non-Disclosure Agreements) as illegal
as other contracts whereby people agree to drop fundamental rights
will be enough to preserve basic human rights regarding information,
and that market pressure will force most people to publish their source code.
Also, the state could demand source code as a customer entity
for all programs used within public administration,
and insurance companies could demand that source code be published
for any and all software used in activities covered by insurance policies.
Publishing source code could thus be part of standard civil conventions,
and have recognition within framing laws for standard contracts;
but it should be considered only as standardization
of an essentially voluntary convention between individuals,
not as a fundamental right.
Note that "consumer protection" as you say, _is_ part of what I consider
useful standardization of voluntary convention by law.
But it's thus subject to discussion and evolution,
not a inalienable immutable right.

Or maybe it is a fundamental right, but of an altogether different nature:
I do believe there is a fundamental duty to acknowledge the truth,
and that this duty covers authentification of information,
acknowledgement of its rightful authorship.
Note that this is a duty towards the public
as well as towards the original author.
One doesn't have a right to lie and conceal who wrote what
about a piece of software or other information,
and must acknowledge precisely who did what about a piece of information;
the requirement that precision be as accurate as possible
yields a requirement that patches be published,
"original" sources being patches from the empty source set
(indeed, the only way to prove that one is the original author
is to show the source and let other people see that it wasn't
copied from another source).
But requiring such an accuracy is stretching this right
a bit farther than natural law allows.
Indeed, there is never 100% accurate account of authorship
(where did you get this precise idea from? who inspired you at what moment?),
and increased accuracy demands efforts from the author
and is a service that should be rewarded as such,
and only demanded in case of judicial conflict.
So again, I think that while publishing source code
should indeed become a standard conventional practice,
it shouldn't be a legal requirement.

To conclude, the end of intellectual property won't make all software free;
it will only make free software whose source is not secret.
And it will make secret difficult to keep, and disclosure easy to demand,
providing an incentive to publish sources before they are otherwise
stolen or decompiled or rivaled.
It won't enforce copyleft, and will make copyleft unenforceable as such;
but it will also create a world where information hoarders
are no more protected by law, and where publishing is encouraged.
Copyleft is using the intellectual property system's weapon against itself,
which is a good thing under the system is dead.
But when the system is dead, we'll have to learn to live without weapons,
as civilized citizens of a policed world.

Best regards,

PS: I'm moving this discussion away from tunes@tunes.org
to cybernethics@tunes.org since it's becoming more political than technical.
Please only include cybernethics@tunes.org in your follow-ups.

[ "Faré" | VN: Уng-Vû Bân | Join the TUNES project!   http://www.tunes.org/  ]
[ FR: François-René Rideau | TUNES is a Useful, Nevertheless Expedient System ]
[ Reflection&Cybernethics  | Project for  a Free Reflective  Computing System ]
In its weak form, Utilitarianism sums up as a requirement
of observational consistency of ethical rules.