🚀 go-pugleaf

In article <e7oR522w164w@mantis.co.uk> mathew@mantis.co.uk (Industrial Poet) writes:
>>     If the author distributes software under the GPL in expectation of
>>     consulting income from users who need extra features, you directly
>>     damage the author's income because the author can't fix your version.
>
>Why *should* the original author be able to make money from something I
>modified and which I am selling?

Why *should* you be able to make money from something somebody else
wrote, and from which you have created a derivative work?

>The GNU approach questions my right to control distribution of my software. I
>question the original author's right to make money from supporting modified
>versions of his code sold by other people.

It is *******NOT******* *your* software... it is the *original*
*author's* software which he has graciously permitted you to make
modifications to, a privilege which YOU WOULD NOT HAVE had he not
granted it to you, under both US copyright law, and international
conventions.

>So the question remains:  Why is reimbursement of a GNU program's original
>author (by means of releasing the changed sources) legally required if I
>*sell* or *distribute* modified GNU software, but not if I *use* modified GNU
>software?

Because that's the way the author wants it!  The author has recognized
that people sometimes need to make changes to software in order that
it will operate in their environment, or to add needed features, or
what have you.  The author says that he doesn't mind if you do this
for your own personal use, but if you distribute a DERIVATIVE WORK
based on HIS PROGRAM (sorry for all this shouting) then you must abide
by certain terms, specifically to pass on all the rights you received
from the original author.  That is simply the way the original author
wants his works to be distributed; if you do not agree with this
conditional license, then you can go rip off somebody else's software
which doesn't have as strong a license.  ["Rip off" may be too strong
in certain contexts, in which an author doesn't mind what happens to
his code.]

Let me see if I can explain this another way.  Let's start from the
beginning.

A programmer has created a work.  He wants this work to be available
as widely as possible *in* *source* *form*, for reasons which only he
may be privy to.  In order to make this possible, but to still have
the work be *useful* to its recipients, he needs to grant some kind of
*limited* license to make derivative works (otherwise, it would never
be allowed to be ported to anything, etc. etc. ad nauseam).  The
nature of those limitations depend on the programer's goals; in our
case, the programmer wants everyone to be able to UTSL.  So, in order
to prevent the derivative-work-makers which he has licensed from
making trivial changes to the program and using that as an excuse to
hide the sources from any recipients, he places the restriction that
derivative-work-makers must release the source.  Furthermore, in order
for those restrictions to be propagated through nth-order derivative
works, he further imposes the condition that all derivative works
which are distributed, must be distributed under the same terms and
conditions that the original work came under.  [This makes many
people, including people who write GPL'ed software, uneasy; however,
it is the only way to propagate the conditions through many layers of
derivative-work making.]

This does not answer the question as to why a commercial firm should
want to use the GPL.  But then, it also does not answer the question
as to why a commercial firm would want to give away free software of
any kind, and that certainly happens---look at X, SRC Modula-3, WAIS,
many i386-specific bits of Mach 3.0, and so on.

I hope this helps...

-GAWollman
Garrett A. Wollman - wollman@emily.uvm.edu

Disclaimer:  I'm not even sure this represents *my* opinion, never
mind UVM's, EMBA's, EMBA-CF's, or indeed anyone else's.