Path: utzoo!dciem!nrcaer!scs!spl1!laidbak!att!pacbell!ames!pasteur!ucbvax!hplabs!hp-pcd!hplsla!jima
From: jima@hplsla.HP.COM (              Jim Adcock)
Newsgroups: comp.lang.c++
Subject: Re: libg++ caution
Message-ID: <6590057@hplsla.HP.COM>
Date: 3 Jun 88 18:54:27 GMT
Article-I.D.: hplsla.6590057
References: <6590055@hplsla.HP.COM>
Organization: HP Lake Stevens, WA
Lines: 60

| For those who prefer to form their own opinions, I have appended the
| text of the license that applies to G++.
| 
| I always find it strange that software developers wish to benefit from
| using freed software, but don't wish to past that benefit on.

I consider the issue of whether certain software tool or library is
"free" to be used or not to be a valid technical issue -- just as much
as a warning that the "turkey" program might distroy your file system is
a valid technical issue to bring up among users who might unknowingly get
their hands on the "turkey" program.

If comp.lang.c++ is not the right forum to discuss possible problems with
g++ and libg++, then pray tell, what would be the right forum ???

Regards the license, the problems seem to primarily center around the
issue of when is an independent software work developed using FSF tools
a derivative of those FSF tools, or not.

If the act of compiling your independent software program using FSF tools
places coded versions of FSF code in the generated results, then there 
might be some claim that your work is a derivative work of the FSF tool.

But most compilers and libraries DO in fact do this.  If FSF wanted 
program writers to be free to use gcc or g++ to compile whatever independent
programs they develop, for whatever ends they choose -- then presumably
FSF would be willing to make a statement to that effect.  We have not had
any luck getting this.  On the contrary.

I personally would be happy to pass on the benefits of "free" software use
to other software developers.  I would happily give some of my own salary
to "free" software developers who develop tools that I am "free" to use.

But in general, I and most of software writers are not "free" to allow
our software development work to be usurped by proponents of "free"
software.  Any software that we choose to place in the "free" domain
must be a conscious rational choice on our part,  not an act of coercion,
nor a "mistake" made on our part as a result of our misunderstanding of
legal licenses.  And in general, we are not "free" to do with our
software development work what we please, since in general we work for
one or another company, which puts restrictions on our actions.  This is 
a choice we choose to make.  Stallman made a different choice.

But surprisingly or not, many of these same companies ARE willing to allow
much of our work to pass into the public domain.  Much software developed
at these companies does become the common property of software developers
everywhere.  Many ideas developed at these companies are published in
public journals, for use by everyone.  Many of these companies, and employees
of these companies donate millions, if not billions of dollars to charitable
organizations, organizations that work for the benefit of everyone.

But these charitable actions must be conscious, rational decisions on our
part, not the result of our mistaken assuptions about various legal licenses.

If I or others screw up in our understanding of these licenses, at the very
least we could lose our jobs.  At the worst we could get personally sued.

And that doesn't sound very "free" to me.

My own opinions only, and not a substitute for competent legal advice.