🚀 go-pugleaf

In article <1991Nov04.171025.23022@infoserver.th-darmstadt.de>, 
schrod@iti.informatik.th-darmstadt.de (Joachim Schrod) writes...

[ ... pointing out inconsistencies in previous articles,
  and requesting clarification ...] 

>In article <4NOV199101065889@minnie.me.udel.edu>, johnston@minnie.me.udel.edu (Bill Johnston) writes:
>> What I object to are the assertions that CopyLefted software can
>> be sold commercially.  This is a case in which GNU is trying to 
>> "have it's cake and eat it too".  

>Can you please cite this assertion? 

I am not sure that I understand what is requested here.  Am I asked to
support the proposition that 1)  GNU claims that CopyLefted software can 
be sold, or that 2)  this claim is made by GNU but (I claim) it's an
example of GNU trying to "have it's cake and eat it too"? 

I'll try to address both possibilities.

To date, my arguments on this point have attracted two types of reply 
from FSF supporters which are somewhat at odds, in my opinion.

The first type of reply is "But of course you can sell CopyLefted
software, that is what we do at the FSF."  The second type of reply,
which also comes from FSF supporters, is that "No you can't sell
CopyLefted software;  the software is free, and any fee that you
pay is a distribution fee, charge for media, etc.

Why this difference?  I believe that there is disagreement among
supporters of GNU, FSF, and LPF over the question of how far to go 
in changing intellectual property laws.  Some want only to eliminate
the possibility of UI copyrights/patents;  others hope to abolish
the entire concept of "intellectual property".  The latter and more
radical group prefers to think of software as something that cannot
be owned or sold, because it belongs to the realm of ideas, like a 
law of physics or a mathematical theorem.

In practical terms, the "moderates" are concerned mostly with the 
effect of UI copyright laws and algorithm patents on their freedom 
to write code.  They may copyleft all or part of their own code, 
but have no problem affirming the right of others to copyright
and sell software and restrict copying in the conventional sense.

The radicals see "copyleft" as a tactic, rather than a fundamental
principle.  The reason for this is that they understand that the 
CopyLeft is itself based on the principle of ownership of 
intellectual property.  

One cannot restrict the use of something that one does not (or cannot) own.
Because the right to Copyleft -- and the right to copyright and restrict
subsequent copying -- arise from the same source, the tactic of "CopyLeft"
is seen as a temporary use of existing copyright law to further the free
software agenda.  The eventual goal is to overturn laws and court decisions
which they claim have manufactured the "artificial" notion of software
intellectual property.  If successful, this would make the CopyLeft
impossible (and unnecessary).  The most extreme expression of this radical
viewpoint extends the overthrow of software intellectual property rights 
to include all intellectual property rights.


The point that I have made -- and it is an expression of opinion -- 
is that there is a deliberate ambiguity in the GNU position on 
"selling" CopyLefted software that is really an attempt to make 
the FSF more palatable to the groups that it stands to affect most,
the professional programmers who make up the commercial software
industry, and the academic programmers who like the freedom made 
possible by publically available libraries of source code.

IMO, this ambiguity is well and fairly expressed in the quote that
I used from Richard Stallman's interview on National Public Radio:

=======================================================================

Q:    When you talk about "free", in the Free Software Foundation,
      how are you using that word?

RMS:  Well, the word "free" has two meanings in English.  One has to
      do with no price.  The other has to do with freedom.  In the 
      term "free software", I am talking about freedom, not price.
      So when a program is free, it means that you the user have the
      freedom to make copies, the freedom to study it and change it
      and improve it.  But it doesn't say anything about whether you
      pay money to get it.  You might get it without paying money, 
      or you might pay money to get it.  At the Free Software Foundation,
      we distribute software in two ways.  We put copies on the 
      computer nets and people can copy that and they don't have to
      pay anything.  Or they can send us money and we'll mail them
      tapes.  And that's how we raise most of our money.

=========================================================================

It is thus possible to REQUEST payment for CopyLefted software,
but one cannot REQUIRE payment.  In the business world, very 
few are willing to pay for what they can get for free.   In the 
"user world", even fewer will pay for things that are free.

Because the amount of money that is available to pay programmers
is proportional to the number of dollars spent to "get" that
software, such a policy pushed to its logical conclusion will
mean that fewer dollars will be available to pay you, the programmer.


To return to Joachim's request for clarification, he asks:

>And can you please stop changing
>your terms? In your former article
>(<3NOV199114383049@minnie.me.udel.edu>) you wrote
> 
>> If you want to guarantee that the
>> code will be unavailable for commercial use, then CopyLeft it.

>The term you changed is `usage.' First, you wrote of `usage,' now you
>write on `selling.' Such statements as `GNU source is off-limits to
>anyone who plans to write commercial software' (which you made in
><3NOV199113581633@minnie.me.udel.edu>) are not usable, because they
>do not differentiate enough.

I used the term "selling" the traditional sense:  "selling" is what the 
baker does when he trades you a loaf of bread for money.  The payment is 
not voluntary;  if you want to have the bread you are REQUIRED to pay.  
What GNU does with software is not "selling" under this definition.

In the sentence quoted above, the term "use" refers to using GNU
source as the foundation or building blocks for other programs.

It doesn't refer to "using" EMACS as a word processor, or "using"
gcc with your own interfaces to create commercial software.  The
CopyLeft does not restrict this kind of "use".

However, the use of CopyLefted source code in a new software 
project does require that the CopyLeft be carried forward into 
the new project.  This means that many programmers will not
be able to use this code as the basis for new work;  not just
those whose want to write commercial software, but any programmer
whose client requires clear copyright in the finished project.
For example, many academics have grants that require that the 
product of the funded project be placeable in the public domain.

Publicly available source code has traditionally played a significant
role in such projects.  Examples are to be found on internet ftp
sites all over the world.  It is a mechanism for preserving and 
sharing the value of one's programming effort that has served us well.

If one opposes the notion of software intellectual property,
the logical response is to refuse to assert rights accruing 
from intellectual property;  one can place one's work in the 
public domain.  This means that EVERYBODY can use it, not just
those whose employers are enlightened enough to permit them to
work on projects that can be CopyLefted.

Some programmers may have the kind of skills and reputation that             
would enable them to choose only those programming jobs in which            
the product can be CopyLefted, but most of us do not have the                   
luxury of dictating terms to our employers.                                   


Joachim asks:

>So, please look at the following statements and think of them/reply
>to them. (And note that they do not only react on your specific
>postings, these are reactions to the bulk of postings you did the
>last week.)

> 1. To `use software' does not only mean that you incorporate this source
>    in your programs. It may also be that you use GNU Emacs for
>    typing, gcc -Wall for compiling, gdb for debugging, and that
>    you sell your created software afterwards. I would be happy if
>    commercial C compilers are at the same stability level as gcc...

> 2. If GNU software is really not usable for commercial companies,
>    why do companies like NeXT, DG, HP, DEC, etc., distribute GNU
>    software? Why is the OSF cc derived from gcc?
>	Please explain this.

I explained this above.  I did employ the term "use" with the word
"source" as direct object;  I specifically referred to the use
of GNU source code as the foundation for subsequent creative work.
Because the CopyLeft is designed to carry over into derivative
works, GNU source can't be used for projects that can't be themselves
CopyLefted.

This is only a problem because it conflicts with one of the stated
objectives of the FSF, which is to promote the distribution of 
software that users can modify and build upon. 

These objectives would only be fully realized in the case of 
the CopyLefted code in the case that the so-called (my term)
radical agenda of some in the the FSF -- the abolition of software 
intellectual property laws -- is realized.

If one believes that this can and should happen, then the CopyLeft
is a logical tactic to further this agenda.  If one believes that
intellectual property can't or shouldn't be abolished, then the 
act of CopyLefting means that the usefulness of your source code is
partly limited for those who can't attach the CopyLeft to their work.

> 3. I agree with you that someone who wants to sell his program should
>    not use the GPL. (After all, his software will not be free.) But
>    please explain me, why is he/she not allowed to use the GNU tools?

I did not make this argument.  The GPL does not restrict the private
use of CopyLefted software. 

> 4. If someone releases his source code under the GPL, he can of
>    course sell an other version of his program. This other version
>    may have a better user interface and other value-added stuff.

The only objection I have made is that everything the GPL does
to ensure that rights to use source code is unrestricted, is 
accomplished more effectively by releasing the source code to
the public domain.  The CopyLeft is added baggage that serves a
political agenda without furthering people's freedom to use
the CopyLefted code.

For those who say, "Yes, but we've got to do _something_
to change the current state of affairs;  user interface
copyrights are poisoning the well...", let me offer a few
comments.

First,  I agree with you.  I have publically called for Apple
to put an end to its UI lawsuits, and did so in an article posted
to this group, and xrefed to the Macintosh groups.  I didn't 
expect that Apple would immediately drop their lawsuit as the
result of an article from one person, but I am on record on 
this point.  

I also oppose the LPF boycott of Apple.  I think that it is 
arbitrary, and that the charges of "terrorism" that have 
appeared in this group have strengthened the hand of those at
Apple who favor the lawsuit.

I have also argued that the LPF boycott has been ineffective.
This is mostly because Mac buyers are either DOS/Unix refugees
who have little interest in text based non-interactive software.
For the non-programmers who account for the lion's share of Mac
sales, the non-availability of GNU software is irrelevant.
The boycott affects only those who are inclined to support GNU,
and does so to a limited degree because we are porting software
on our own, in the same was that MS-DOS users are.


Then what would be an effective strategy, assuming that we
must take action to counter the negative impact of UI copyright?


I suggest that someone, or some group should anonymously write
a free Macintosh-like UI to work in conjunction with the GNU OS.
One assumes that the GNU OS will not impose a standardized
GUI as does the ROM-based Macintosh ToolBox, but I imagine that
it will support a plug-in UI.

This could be done as an act of civil disobedience, and as a test
of the hypothesis that Apple Computer is out to sue "anyone who
comes up with a reasonable DeskTop program".  In order to limit
the work involved, the focus should be on recreating precisely
those elements of the Mac GUI that Apple claims to own in its
lawsuit with Microsoft and HP.

Then see what happens.

-- Bill (johnston@minnie.me.udel.edu)