Re: [dvd-discuss] Gedanken Experiment -Unix and Norton

["John Zulauf" <johnzu(at)ia.nsc.com>]

Seth Johnson wrote:
> 
> John Zulauf wrote:
> >
> > Seth Johnson wrote:
> >
> > > The application of the trade secret model is far better because it separates
> > > private interest from exclusive rights.  "Publishing" does not simply mean
> > > "readable" or "viewable."  It means making available to the public.
> >
> > Publication != "freely given" however.
> 
> The only restriction on published information is the reproduction of
> original expression.  I don't understand what you mean by "freely given."

Sorry.  I was confused by your phrase "making available" -- I thought
meant fully available in the "public domain" sense.  'Publication !=
"free given"' was shorthand for "just because I publish something
doesn't mean I have to give it away."

> If you mean to imply that the published work is not
> open to free use, you're wrong as far as anything besides reproduction of
> original expression.  

It's not open to "free" (as in unrestricted) use.  It is open to
ordinary (reading and studying, etc.) and "fair" use.  The boundaries of
that "fair" are marked by the four part test of Harper and Row v.
Nation.

1. the purpose and character of the use; 
2. the nature of the copyrighted work; 
3. the substantiality of the portion used in relation to the copyrighted
work as a whole; 
4. the effect on the potential market for or value of the copyrighted
work. 
 

> If you mean things won't be given for free, it's not
> relevant to the point that publishing means making information available
> such that it can be used freely, save for reproduction of original
> expression.  

You are making my point.  If the copyright holder cannot prevent
extraction and reimplementation of a "critical invention" or
"differentiating technology" exposed in published code, he/she/they will
strongly oppose any requirement to publish that code.  Thus without
means to prevent this, non-free source code will not be broadly publish
and we are back to the list of horribles we have today.  The only means
to prevent the "making" of a software invention is by a patent.

QED.


BTW. This the second time you have used "reproduction of original
expression" as the only exclusive right.  This is clearly not true. 
Translations (from english to german or C++ to C# or Java or pig-latin
...) are not reproductions of the "original" expression, yet are within
the exclusive rights of the copyright holders.  Derivative works (such
as adding a few methods to some publishing object code) certainly are. 
My point is that publication withholds all these rights still.


> You can't support your point with the false notion that the
> economic incentive has priority over the purposes of the exclusive rights
> clause, or the intrinsic nature of information as such.
> 
> > As an example, a copy of the
> > source (or the right to one) could accompany each binary copy of the
> > application and fulfill a publication requirement.  However, if
> > algorithm can only protected as a trade secret
> >
> > (a) The secret is never shared (and never falls into the public domain)
> > and potentially lost (c.f. the Stradivari (sp?) brothers and their
> > irreproducible violins)
> >
> > (b) a source publication requirement will never be accepted by software
> > companies
> >
> > (a) and (b) tending to be antithetical to "promote progress" I'm not too
> > fond of them.
> 
> If you want to incentivize disclosure, figure out what might encourage that
> without violating the intrinsic freedom of information as such.  It doesn't
> follow that software should be patented.

I don't see any other options to encourage disclosure.  Having written
software for many years, I cannot believe that anything short of patent,
or patent like protections of key algorithms would incentivise
publication of company source code.  Politically, mandated publication
is a non-starter.  The populace doesn't understand it and a broad
coalition of commercial interests would violently oppose it.  The only
acceptable (to non-free software developer) compromise would include
protection for the software inventions implemented within the source.

If you want in incentivise non patenting (or freely licensed patents) by
some sort of tax credit, deduction or the like go ahead an propose.  But
without first getting to a publication mandate that would be acceptable
politically, you won't have anything to work with in the first place.

> 
> Overextending exclusive rights such that you restrict the free use of
> information as such is an expression of allowing the private interest to
> overtake the public interest in exclusive rights policy -- not just the
> "promotion of progress," but fundamental rights.

How is extending patents to software invention any expansion
(philosophically) of the current policy of patents for tangible
inventions?  Unless your argument is that we should have no patents, the
above paragraph seems highly exaggerated.

> 
> Nothing has changed as far as what governs the choice to keep a discovery a
> secret or to publish it.  There has only arisen an egregious desire to lay
> inappropriate claims on new media and digital tools.

One important thing has changed.  One of the most important elements of
"science and the useful arts" (algorithmic development) currently has no
means but trade secret to protect it.  Thus the decision whether to
publish is a simple one for a "for profit" entity.  Publish and lose
economic advantage or hold secret and keep that advantage.  Without
protections for this class of inventors, we go back to the "bad old
days" before patents when all there were was trade secrets, without
publication of useful technologies.