Re: licensing issues

"John van V" <[email protected]> wrote:
>
> > The dual license is already such a compromise.  What's wrong with the 
>dual
> > licensing scheme?
>
>Ok, I'm learning here, please send me the link.

Ships with Perl.  Perl is copyrighted and the copyright
holders say you can use their copyrighted code under
either of 2 licenses.  Browse your local Perl install
for more.

> > Well, this obviously isn't true in general since Perl is a project to
> > create a programming language and GNU is a project to create an 
>operating
> > system.
>
>Actually, this the ~only~ obvious thing here.  What I
>just learned from the GNU/FSF/UWIN/MinGW issue is that
>perl ~is~ legally defined as an operating system.

Defined by who?  I am curious here.

>So we are all learning here.  I am in the perl business,
>and I know that litigation will kill my business.
>
> > If you meant that the people working on the Perl project and the people
> > working on the GNU project have the same licensing goals,
>
>Have the same overall economic and technological goals,  I
>am saying that the licensing differences should not impact
>the creation of free s/w.

They don't have the same economic and technological goals.

Larry Wall wants to help people get jobs done with a mimimum
of inconvenience so long as nobody tries to claim his hard
work as theirs.  The FSF wants to destroy proporietary
software.  Those goals both result in free software but are
not the same and in many key respects are not even very
similar.

Very specifically, the FSF actively wishes to inconvenience
people who are developing proprietary products.  Larry wants
to be nice to them by default.  This is a huge philosophical
divide.

>Also, I am speaking for others here, who are on this list but
>are not speaking up.  Not me :)

Let them speak up then.

>The perl XSUB process pulls in GNU code and glues into the
>perl batch system, allowing a single point of entry,
>preventing bloat by sharing all the tiny pieces between
>packages.  Ultimately I would like to see these fully scoped
>modules travel thru the ether to any type of  perl VM.

As soon as you have mixed GNU and Perl in a binary, the only
license that you can use is the GPL.  For Larry to accept
this is is contrary to his goals.  For Richard Stallman to
give Larry an exception is contrary to his goals.  Unless
you understand that the licenses reflect the actual goals of
actual people, you won't understand license conflicts.  As
long as the people continue to disagree, the license wording
is a red herring.

>So perl is bigger than GNU because it is no longer glue but
>moldable resin structure.  This will be true in the public
>perception as well.

I doubt you understand what the GNU project is.

The objective of the GNU project is to create a world in
which proprietary software has been completely replaced by
free software.  Creating an operating system, utilities,
libraries, and so on are necessary to the overall goal, but
the goal is all software, not just some.

>Ø so is X, TeX, and any number of other pieces of software that
>Ø well, maybe our efforts can achieve an overall agreement that Ø keeps the 
>licenses from conflicting.
>Ø
>Ø aren't even under the GPL.
>
>Ok, maybe they should be under the GPL, is there any reason
>we cant discuss what such a document would look like with
>the possibility of making it internation law ??

Why should they be under the GPL?

The stated objective of the X project was to get their
software used.  They actively wanted proprietary companies
to borrow their code and use it.  The GPL would restrict who
could reuse their code, and that violates their goal.

The stated objective of TeX is to create a definitive piece
of software for taking text input and producing well-defined
output.  TeX is not intended to be modified with the result
possibly polluting the standard.  The result is that the
exact same ASCII document defines an attractive output
(very literally of the same quality you would expect from a
professional publisher, with the output document specified
to the visible wavelength of light) and does so on every
single operating system.

If you need to create an archive of technical documents this
guarantee of protection from bitrot is very important.  And
that is why other formats (eg variations on PostScript and
Microsoft Word) have made no inroads against TeX in
mathematics.

Again, the GPL's guarantee that you can change code makes
very little sense for a piece of software that is not
supposed to change and which has had no bug discovered since
1986.

>No time like the present.

Memorize this.  Licenses reflect goals.  In the real world
license conflicts usually are the result of conflicting goals.
To the extent that goals truly oppose each other, the
licenses cannot be brought into agreement.

> > What I think you were trying to say here is that the people working on
> > Perl and the FSF (which is a different entity than the GNU project) have
>
>Once again, I don't believe in licensing, I just
>discovered that I have a talent for negotiating the issues
>If I can help, I will want to.

Before you can negotiate the issues beneath the licenses
you need to understand the goals of the people involved.

>Speaking for "most perl people " is a dangerous thing to
>do.  Speaking to people is how I got interested in this
>issue.  What I have learned is taht people are not
>comfortable w/ the schism between all the different
>licenses and a few are downright disturbed.

Good.  The schisms are bad.  They result in duplication of
effort.  They inconvenience people.  A significant fraction
of developers who want to develop free software do not care
about license issues.  And continue to not care until they
see how it affects them and people who want to use their
work.

> > working on Perl seem to be at least reasonably content with Perl being
> > used for proprietary projects if people wish to, are interested in 
>finding
> > ways for proprietary software companies to work more closely with the 
>Perl
> > community and to be able to write Perl modules and the like, and 
>otherwise
> > are not particularly strongly behind the idea that all software should 
>be
> > free.  There are exceptions, of course (it's a large community), but the
> > FSF has a much clearer political goal.  Perl doesn't really have as much
> > in the way of political goals.
>
>I guess the desire is to allow GNU s/w to pass thru to the
>perl license w/o penalty, somehow.

Richard Stallman very strongly does not want that to happen
and has written a number of essays laying out exactly what
he believes and why he would never agree to what you are
suggesting.  If you try to deal with Richard and the FSF
with the belief that their goals have any relation to your
goals, you will fail to convince them and be baffled at why
you failed.

If you understand their goals you will still fail to convince
them to do things they don't want to do, but at least you
won't be so baffled. :-)

>Also, the GNU/UWIN issue is relevent becuase we need a legal
>way to compile and distrubute Win32 perl, plus I just became
>friends w/ the whle
>cc: list :)

You can compile proprietary software with gcc.  While you
might have distribution problems with gcc itself, you should
be able to compile and sell things with it.

>That is what I am trying to discover in this thread.
>
>Any hoo, licensing by design is a high noise / low signal
>topic.  I was ~not~ my idea I wish I was as good at coding
>as activist politics :)

Good luck,
Ben
_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com