Re: [Patents] confusion about terminology

[Hartmut Pilch <phm@a2e.de>]

> >>We must be careful to speak about "software publisher" as intellectual
> >>works (such as computer programs) that are published and not manufactured.
> 
> >Yes, we try to avoid that.
> 
> You are wasting your time with this tactic of relying on Berne to help
> you fight software patents.  First, Berne is a bit vague on wht forms
> of protection literary works should be provided - Berne mostly justs
> says that they should have protection.  Protecting software with patents
> is not incompatible with Berne.

If software is to be "protected as a literary work" (Art 10 TRIPs), then
it falls under the Berne regime and not under the patent regime. 
 
> Even if you assume Berne mandates copyright protection for literary
> works, and you can somehow convince everyone that software is any more
> literary than a circuit diagram, 

Software is about as literary as an instruction manual.

> you have the problem that Berne has no copyright laws as part of it -
> instead, you have to defer to each country's (or regions - EU) copyright
> law.

> And the governing copyright law that has plagued software caselaw for
> decades to the point of rendering software copyright pretty much useless
> for any aspect of software beyond identical copying, is that only the
> expression of "things" covered by Berne are protectable, not the "ideas".
> That is the foundation principle of American caselaw and the WIPO
> copyright treaty.

Software copyright is not useless from a practical point of view at least.

The "Harmony" project tried to rewrite the Qt library, which was published 
with source code under a nonfree license, but Harmony never got very far 
because they had to keep some distance to Qt.

In combination with source code secret, copyright is even more useful.

See also

        http://swpat.ffii.org/papers/bgh1-bs90/

for a model copyright case where non-patentability of software is also 
explained as a consequence of copyrightability.
 
> And for thirty years, few if any have been able to determine an
> objective way to distinguish "idea" from "expression" in software
> - read US court decisions to see lawyers and judges failing to do so.  

Objective tests are not always needed for laws to function well.

> Indeed, for general literary works, like books and movies, the courts
> are struggling to figure out how to distinguish idea from expression.  
> It is a real intellectual mess.

As is the case with many legal notions which are nonetheless useful.

It's different when you use a notion such as "technical" without clear 
definitions as a pretext for patent offices to impose far-reaching 
restrictions on people.  The restrictions imposed by copyright are not 
far-reaching.  That is why fuzziness in legal concepts is easier to 
tolerate there.
 
> But at least for software, software is pretty much process and method
> which is comfortably compatible with a patent law that has no
> idea/expression dichotomy, which is why over the last fifteen years
> most serious software protection has switched over to patents.  You
> may not like it, but it is far less illogical than reasoning about
> software in the copyright world.

Patent law is full of illogics (Sir Arnold Plant and Fritz Machlup were
very good at pointing it all out, and for software patents Newell "The 
Models are Broken" of 1986 did an excellent job at dissecting the logical 
mess of patents).

Still the question of whether patents or copyright is more appropriate is 
not a question of which is more logical.  An illogical law can be 
appropriate, a logical one can be inappropriate.
 
> So this Berne tactic is a waste of time - since the end result of using
> Berne lands you back in the dilemma of distinguishing idea from expression
> in software, which on both sides of the Atlantic has been an intellectual
> quagmire of conflicting opinions for thirty years.

> Also, it is silly to say that software is not "manufactured" to make it
> more literary, when increasingly more and more of what is literary
> (especially music) is manufactured (the insulting Hollywood term for
> this is "formulaic").

Newspapers are also produced in factories. 
That doesn't make them industrial products.

It is as feasible and meaningful as ever nowadays to distinguish between 
information & service goods on one hand and industrial goods on the other.

Industrial in the sense of patent law means something like material or at
least requiring a commercial effort of bringing a tangible product to the
market.

This is regularly not the case with information goods or services.

Even the oldfashioned newspaper can be produced in an office and then 
handed over to a printing factory which is ignorant about the content of 
the newspaper.

The European Parliament voted for an article 2b that defines "industry" as 
"automated production of material goods" for those reasons.

This whole debate is quite old and making little progress.  There is still
a need of updating pages like

        http://swpat.ffii.org/players/aharonian/

and the addenda wiki

        http://kwiki.ffii.org/SwpataharonianEn

so as to avoid a sisyphus-like eternal repetition.  I'd even welcome
Greg to try to create the FAQ with us in the Wiki.

-- 
Hartmut Pilch, FFII & Eurolinux Alliance              tel. +49-89-18979927
Protecting Innovation against Patent Inflation       http://swpat.ffii.org/
300,000 votes 2000 firms against software patents    http://noepatents.org/