Hi,
Lots of flames and hearsay from both sides, but also
some very valid arguments.
I think we should try to find out exactly where we
agree and where we disagree. This discussion is too
important to use for another flamewar about licensing
ideologies.
We can both agree that neither of us want to violate
the copyright of the other part, and that each copyright
holder has a right to distribute _his_ copyrighted
works under any license _he_ chooses.
I also think we can agree that both jBoss and Tomcat
are independent products that are both useful without
the other part.
(Just joking here) We have:
usefulness(jBoss)+usefulness(Tomcat) <= usefulness(jBoss+Tomcat)
(synergy), and jBoss+Tomcat == (License problems).
But usefulness(License problems) == 0, so unless
we get this sorted out we have to derive:
usefulness(jBoss)+usefulness(Tomcat) <= 0.
;-)
Peter Donald wrote:
>
> >| I think it would definitely be safe to download a set of RPMs (one
> >|per product) and then install them all and configure them to point to each
> >|other (using network protocols, standard interfaces, etc.), but I think
> >|it's very questionable whether you can put them in a single pre-configured
> >|package.
> >
> >explain it to RedHat,
> >This is turning silly
>
> RedHat complies. None of it's RPMs contain GPL and GPL incompatable
> products. They were blasted a while back because they broke this with one
> of their packages thou so I don't think they will make same mistake again.
> What red hat does is distribute a medium with multiple packages.
I have this strange feeling that you both agree on this, but that
none of you want to admit it...
RedHat _does_ distribute software with GPL license and GPL-incompatible
license on the same physical media.
We can probably all agree that GPL does allow this due to the last
paragraph of section 2: "In addition, mere aggregation of another work
not based on the Program with the Program (or with a work based on the
Program) on a volume of a storage or distribution medium does not bring
the other work under the scope of this License."
What RedHat found out they could _not_ do was to combine software with
GPL license and GPL-incompatible license into the same binary package,
and then distribute this package.
So I guess that the question boils down to: When do we have "mere
aggregation" and when do we have "combined software" ?
If I burn a CD with jboss.jar (as distributed from the jBoss site
under GPL) and tomcat.jar (as distributed from the Tomcat site
under APL) there should be no problems, as this is "mere
aggregation". After all both distributions are seperate, except
for the fact that they have been placed on the same physical media.
It might become a little more tricky if I decide to order
a batch of these CDs at a CD production facility. They want
me to send an ISO image of the CD by email. But the ISO image
is a single file and it contains both jars. Mailing the ISO
image is distribution in the GPL sense, so _if_ the ISO image
is "combined software" I would break the APL license.
But is this ISO image "mere aggregation", or is it "combined
software" ?
There is a good argument in favor of "combined software":
- The ISO image is a single file, so the software must have
been combined rather than aggregated.
There is another good argument in favor of "mere aggregation":
- Neither the jboss.jar nor the tomcat.jar have been changed
or modified in any way. They are both embedded in the same
file, but they have not been combined.
How do you think these two conflicting arguments would hold
in a court of law?
> if tomcat is contained in same archive then it has to be GPL.
This is similar to (though not the same) as the ISO image
case above.
If an archive containing unmodified jboss.jar and unmodified
tomcat.jar is distributed under GPL, that would clearly be a
violation of APL. And distributing it under APL would be a
violation of GPL. But do we have to distribute the archive
under one of these licenses?
If both jboss.jar and tomcat.jar are unmodified, I guess that
it would be possible to argue in favor of "mere aggregation"
and claim that the archive containing them is simply a volume
used for distribution. After all, they are both unmodified and
neither jboss.jar nor tomcat.jar can be used until the archive
is unpacked.
Both GPL and APL allow distribution and neither are viral in
case of simple aggregation, so it _should_ be possible to
distribute an archive that simply aggregates the unmodified
software from both of us. Do you agree on this?
I think it makes sense to speak about _unmodified_ software
only, as none of us are interested in any code forks.
> If dynamically linked via configuration file through a standard interface*
> and then there is some intermediate code that links to interface then that
> is OK.
>
> To do this you need to supply 3 archives.
> * jBoss archive (under GPL)
> * Tomcat archive (under APL)
> * linking layer (under APL and GPL compatable license - usually public domain)
Now this is productive: Pointing out a possible solution.
I am not using Tomcat so I may be wrong on this, but I
think that the general idea is to make jBoss independent
of the actual web application server used. Instead of
directly calling Tomcat to initialize for embedded
operation, some (end user editable) configuration file
entries are used to hold the name of a class to be loaded
and a method to be called for initializing the embedded
web server used, with fallback to "no web application
server" if initialization of the web server fails. As
most jBoss developers agree that Tomcat is the best web
application server around, it is intended that the default
configuration files for jBoss should contain entries for
starting Tomcat.
Do you think that the name of a Tomcat class and method
as defaults in an end user editable configuration file
would be a violation of the Tomcat APL license, or do
you think this would come under the "fair use" clause
of the copyright legislation?
> >1- we have a LICENSE gentlemen, words, black on paper... That's what we work
> >from.
> >And the words are clear, I really honestly don't see why the big confusion.
> >RMS could be a "auto-response-program" that spouted "all must be GPL" that I
> >wouldn't care,
>
> Then would you mind if I emailed him and informed him of the situation ?
Before taking his time we should at least find out exactly
where we agree and where we disagree.
But please note: What counts here is the legal meaning of
the texts of GPL and APL, _not_ the strong political views
of an idealist.
<personal-opinion flamebait="no">
RMS is an idealist. He thinks that _all_ software should be
free and that copyright should not apply to software. The
GPL license does not exactly reflect his opinion, rather it
is a compromise between his opinion and what is possible in
the real world. When this compromise turned out to be too
close to RMS' opinion in some cases, the Library GPL was
created. RMS always disliked the Library GPL and has even
tried to write a new Lesser GPL to replace it, but with
little luck.
I fear that by asking RMS we get his personal opinion
rather than an unbiased legal interpretation of the GPL.
</personal-opinion>
Best Regards,
Ole Husgaard.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
