Brave GNU World, Issue #5

Georg’s

Permission statement below

Issue #5

[DE |
ES
| FR |
JA]
Welcome to Georg´s Brave GNU World. This issue is dedicated
mainly to the legal issues of free software, because politicians
worldwide are struggling to create laws for the “Information Age”,
and lately they have been following a dangerous course. Examples of
this are the developments in Australia and the plans for patentable
program concepts in Europe. Even if the arguments in this column
are very much based on Europe, the facts are the same worldwide and
should be interesting for everyone.

Licenses and Patents

To explain why the planned patentability of program concepts is
not only useless but dangerous, its probably best to start at the
“Communication from the Commission to the Council, the European
Parliament and the Economic and Social Committee” that can be found
on the internet [4].

This communication blames legal problems with the prosecution of
so-called “illegal copies” on the lack of patent protection for
program concepts and concludes that the internal market is being
weakened by the lack of patent laws.

The comparison of literary work and source code made in the
above study is not uncommon as it helps to describe a new medium
with old terms. Although the comparison may have its weaknesses, it
is more realistic than most people realize. The stylistic range
that programmers show is not at all unlike the range of literary
authors. This is one of the reasons why it makes sense that source
code legally meets the standards of literary work, which by
definition doesn´t need to be patented as it is protected by
the copyright. By the way: this only becomes clear after reading
footnote 6 to paragraph 3.2.2. In the text itself it says
“patenting of programs”, the footnote clarifies that patents can
only relate to the idea behind the program.

The paper´s reasoning tries to show a need for patents by
pointing out the current problems with prosecuting violations of
copyright laws; the commission then suggests that the patenting of
program concepts should be allowed as long as they present a
technical novelty. Applied to the book analogy this means that
while books themselves are subject to copyright, the commission
demands a patent law for entire literary genres.

It is the declared goal of politics to help innovation. Whether
it would have helped innovation in literature if, say, George
Orwell had claimed a patent for “books that describe a repressive
regime” is doubtful.

If the parliament follows this recommendation and establishes
such a patent law, it´ll become necessary to have courts
decide about possible violations of this law. Since the spectrum of
solutions for related problems is very wide due to different
programming styles, even hardcore computer professionals sometimes
can´t say whether something is a new or reinvented concept.
This makes the salary of the lawyer the all-deciding factor in such
a process, inevitably harming small firms and entrepreneurs while
supporting the creation of monopolies.

The USA already have patent laws like this and it is possible
that it is another reason why firms like Secure Shell [5]
emigrated to Europe. That Microsoft is highlighted as a positive
example for the use of patents doesn´t need further comment
in the times of the antitrust trial.

It is an essential characteristic of software that the current
state of the art creates new problems that are being realized and
solved by different people at almost the same time. This usually
leads to a struggle between programs, which only one or two
survive. The competition leaves the survivors with higher quality
and better adapted to the users’ needs. This parallel evolution is
a major factor for a stable and healthy information
infrastructure.

The copyright laws already create a lot of separation and
distrust but the introduction of patentable program concepts would
have a much bigger effect. A lot of resources are wasted by marking
off and monitoring the competitors or are consumed by repeatedly
solving standard problems. Considering the lack of computer
professionals this waste of resources is a significant hindrance to
innovation.

A lot of big firms have realized the advantages of Free Software
and open development recently. The IBM Deep Computing Institute,
for instance, has put the source code of its Data Explorer
visualization software online [6]. The
increasing quantity of commercial Free Software shows that the
business world is adapting to the needs and workings of a new
medium. Trying to support this with a legal framework is a
praiseworthy idea, but a patent law for program concepts is a step
in the wrong direction.

The “computer medium” follows different rules than the
traditional industry. Establishing such a patent law would merely
be hiding the symptoms; the question should be how to fit the
business concepts to the new age and not how to preserve old
concepts at all costs.

You´ll find more information on the homepage of the FFII
[7], i.e. the homepage of the
Eurolinux Alliance [8]
which is a joint project of AFUL [9] (a French group) and the FFII.

Even if this has exceeded the normal size of my articles I hope
the danger of the planned patent law has become obvious now. Please
pick a politician of your choice and send a letter and/or a copy of
this column to him or her. Another possibility would be to contact
the above groups.

But now on to some more practical topics. I´ll begin
with

GNU Smalltalk

is the GNU implementation of Smalltalk-80 [10]. Smalltalk is an
object-oriented programming language with a rather unusual syntax;
details can be found in the relatively complete FAQs on the
Smalltalk webpage [11].

Now that the 1.1.5 release of GNU Smalltalk is already 4 years
old, the new maintainer Paolo Bonzini announced that version 1.6 is
about to come out. This new version will contain a more stable and
up to 5 times faster interpreter, a more complete smalltalk syntax
and a better class library. Now the excellent portability extends
to user-interfaces as well and it´s possible to load
C-modules at runtime.

Plans for the future contain a GNU Smalltalk scripting level
similar to that of Tcl, Perl or Python and extending the
functionality to TCP/IP sockets and regular expressions.

I´ll continue with a program for visualisation of
functions and data.

Xpplot

by Pavel Pokorny from Prague is very easy to install due to the
fact that it consists of just one C source code file [12].
If this program has any underlying design concept, it´s
simplicity. There are just two pages of help text and
self-explanatory functions for xpplot interesting
functionality.

Xpplot can display growing datasets like tail displays growing
logfiles to visualize the development of a running project. On a
mouseclick it outputs the selected coordinates so if xpplot is
being used in a pipe you can easily give points of interest to
another program. Additionally xpplot is script-capable and supports
postscript output.

The program´s roots are deep in science as it was written
in 1991 for the numerical analysis of nonlinear dynamic systems. As
the program is being published under the GPL, it is no problem to
use it for your own projects.

The author intended to create a library version of xpplot and
include more numerical features like vector and matrix operation or
numerical integration although the timing is still unclear.

I´d like to close with a short notice from Finland, which
again seems to be ahead of things. I´ve been contacted by
Mikko Markus Torni with the news that he is currently in the
process of setting up the first GNU user group. Anyone who is
interested in participating or maybe sponsoring this group is very
welcome to send him email [13].

Of course, I can´t miss the obligatory reminder that this
column relies on feedback, so don’t hesitate to send ideas,
questions and comments. You´ll find the address in the info
box [1].

Copyright (C) 1999 Georg C. F. Greve, German version published in
the Linux-Magazin

Permission is granted to make and distribute
verbatim copies of this transcript as long as the copyright and
this permission notice appear.