You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@tomcat.apache.org by marc fleury <ma...@telkel.com> on 2000/10/28 07:10:02 UTC

RE: [jBoss-Dev] Re: jboss on tomcat update

| but at the same time, you have a problem with the GPL being
|viral so you give exceptions for people to use JBoss. Instead, what you
|should do is probably be using the MPL license which will solve your needs
|without having to constantly grant exceptions to people.

???

what 'exceptions'? we never granted 'exceptions'.  Please explain.

|It is funny to me how you say that you are integrating our code which I
|think is great, but the real issue is that we can't integrate YOUR code
|because you choose to use the GPL license.

why not?  what exactly prevents you from integrating our work?  Please be
explicit,

let's not work from hearsay and "impressions" of the GPL, the GPL is very
explicit.

regards

marc



|
|Sigh.

????

|
|-jon
|
|
|


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Peter Donald <do...@locus.apache.org>.
At 02:06  28/10/00 -0700, Jon Stevens wrote:
>on 10/27/2000 10:10 PM, "marc fleury" <ma...@telkel.com> wrote:
>
>> | but at the same time, you have a problem with the GPL being
>> |viral so you give exceptions for people to use JBoss. Instead, what you
>> |should do is probably be using the MPL license which will solve your needs
>> |without having to constantly grant exceptions to people.
>> 
>> ???
>> 
>> what 'exceptions'? we never granted 'exceptions'.  Please explain.
>
>The exceptions that you are granting is by allowing people who write EJB's
>for your server to allow them to not require them to be GPL'd as well. That
>is clearly an exception to the license. This is very similar to what Linus
>has done with Linux and binary kernel modules.

strictly speaking this is not an exception. EJB is a clear defined spec
that has easily replacable components, loaded via configuration files etc.
I can go through the details that make this not apply if you want but you
can trust me this is one of the casses that GPL won't cover all code in one
JVM. Discussion about this has occured in various places (classpath list -
a free LGPL'ed class library) but it basically came down to one thing. As
long as you don't use anything specific to jBoss in an EJB then the EJB
will not have to be GPLed. 

>> |It is funny to me how you say that you are integrating our code which I
>> |think is great, but the real issue is that we can't integrate YOUR code
>> |because you choose to use the GPL license.
>> 
>> why not?  what exactly prevents you from integrating our work?  Please be
>> explicit,
>> 
>> let's not work from hearsay and "impressions" of the GPL, the GPL is very
>> explicit.
>
>I write code for the ASF under an APL 1.1 license. The GPL and the APL 1.1
>are not compatible licenses and it is "illegal" for me to include GPL code
>within an ASF project. Period. Thus, I cannot take JBoss and include it with
>the Turbine Developer Kit because you have things under a GPL license.

right and same goes with GPLed rojects.

>Sigh, I feel like I'm repeating stuff to you again.

;)

You should see ant lists where the same statements were said over and over
again ;)
Cheers,

Pete

*------------------------------------------------------*
| "Nearly all men can stand adversity, but if you want |
| to test a man's character, give him power."          |
|       -Abraham Lincoln                               |
*------------------------------------------------------*

RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by Peter Donald <do...@locus.apache.org>.
>> I am sorry, I should actually provide some information.
>> 
>> We use the GPL to protect the kernel.  The virality of the GPL applies to
>> the "derived work" or "modified work as a whole" of the kernel.

ummm - hello ? - you should seek legal advice as this is NOT what the GPL
saids. It was designed to force developers to GPL their work. See the
philosophy pages at gnu if you think otherwise.


>> Tomcat is not "derived work" of jboss, clearly, wouldn't you say? :). The
>> "modified work as a whole" done in jboss to integrate the Tomcat jar is the
>> MBean adapter (for JMX), the Tomcat Interceptors (classLoaders), and the
>> J2EE deployer that we have developed.  Those are GPL, as per the GPL
derived
>> work virality.

umm - but you link against JMX/Tomcat/JMS/whatever and that is definetly
not legal. They would have to be GPL unless they fall under clause 3.
However  clause 3 has been determined to only apply to libraries that come
with a core java platform (ie j233-j3se-Java Personal Ed etc) and thus you
are in clear violation of the GPL and the law.

Send mail to licensing@gnu.org if you believe otherwise. RMS will usually
reply within 2-3 days after seeking legal advice (if necessary).

>> The GPL applies to derived work in distribution.  Our distributions are GPL
>> kosher.
>> Please don't be afraid of it, and feel free to discuss it...

if it is available it is distribution. If I can get access to it then it is
distribution. I can get access to CVS - thus that is distribution and rules
apply.



Cheers,

Pete

*------------------------------------------------------*
| "Nearly all men can stand adversity, but if you want |
| to test a man's character, give him power."          |
|       -Abraham Lincoln                               |
*------------------------------------------------------*

Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Peter Donald <do...@locus.apache.org>.
At 08:55  29/10/00 +0100, Rickard Öberg wrote:
>Jon Stevens wrote:
>> 
>> on 10/28/2000 5:22 PM, "Peter Donald" <do...@locus.apache.org> wrote:
>> 
>> > Once RMS finds out
>> > about the project misusing the GPL he will start advocating all the GNU
>> > peopls stay away from it.
>> 
>> Someone want to send an email to licensing@gnu.org recommending that RMS
>> take a look at how the GPL is being used within the JBoss project?
>
>I think that would be a good idea. While I understand Marc's frustration
>with these silly things, I also understand Jon and Peters position that
>it really doesn't care what *we* think. The license isn't for us. The
>license is for *other* people who want to help/use our code. And if
>*they* interpret things another way, then that's what counts. Sad but
>true.

I would say it matters little what either you or others think but instead
what matters is the law. As it stands now everyone who has partial
copyright on jBoss code is liable. Any buisness who adopts and tries to
deploy jBoss as a software vendor (rather than a service vendor) accepts
partial liability. Now the "interpretation" I have heard spouted here is
wrong. What you are doing now is illegal - no question about it. 

If you can live with that - fine more power to you. However I feel it is
very very wrong to hoist the product onto people who don't know that what
you are doing is braking the law.

I don't know if you have saught legal council (thou if you have fire them
immediately) and to be honest I really don't care what you do. It would
have been nice to collaborate but I have since worked around what I would
have liked to use from jBoss - thou there is bound to be people who would
like to integrate with you so it may be worth it to aid them.

However the thing that I find really unacceptable is the vicious attack
against the integrity of the GPL. You undermine the tenets of it and you
de-value it. I consulted to a company a while back who wanted to "GPL"
their code. Like jBoss they did not want to limit themselves and wanted to
violate the GPL. Their reason for thinking they could do this ? Because a
project on the net did it. That is the real cost of your actions. Misusing
the GPL like you do is a crime and is considered to be *worse* than
proprietry software. It devalues what FSF and GNU has been working on for
so long.

>A license is a set of words. Words are just to communicate. If the one
>being communicated with do not understand, the set of words used to
>communicate are wrong, no matter how right they are.

In this case it is not the case - it is the one communicating that does not
understand. Simplified the GPL forces all code in the JVM to be GPL'ed with
3 exceptions. The exceptions being

* Dynamically generated code (because it does not make sense to GPL this)
* Code that is part of platform (covered by GPL clause 3)
* "Hosted" code.

Hosted code is a fuzzy term but it generally covers things like jBoss
"hosts" EJBs, tomcat "hosts" servlets etc. This has not been directly into
the GPL but is based on interpretations of lawyers who specialize in this
sort of thing. 

Now you could try to claim that tomcat is "hosted" in jBoss but that would
never hold up in court. 

Now just say your group annoyed someone who used your product - if a
developer happened to live in Australia (my home) they could be put in jail
for 10 years. I expect similar durations in other countries. Everyone who
contributes to jBoss is subject to this.

Now you may think that I am 
a> lieing/mistaken or 
b> a fanatic about some other license. 

Well neither in both cases. IANAL but I have consulted a lawyer in regards
to another similar situation but I have dealt with a lot of siimlar
situations in past. For a recent example you can see the ant-dev mailing
list archives at jakarta.apache.org. 

In regards to b> - be assured that I like the GPL. I used to GPL a lot of
code when I worked with c/c++ and I am going to try and convince whoever I
have to at Apache to make APL GPL compatable (currently one clause makes it
incompatable). 




Cheers,

Pete

*------------------------------------------------------*
| "Nearly all men can stand adversity, but if you want |
| to test a man's character, give him power."          |
|       -Abraham Lincoln                               |
*------------------------------------------------------*

Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Rickard �berg <ri...@telkel.com>.
Jon Stevens wrote:
> 
> on 10/28/2000 5:22 PM, "Peter Donald" <do...@locus.apache.org> wrote:
> 
> > Once RMS finds out
> > about the project misusing the GPL he will start advocating all the GNU
> > peopls stay away from it.
> 
> Someone want to send an email to licensing@gnu.org recommending that RMS
> take a look at how the GPL is being used within the JBoss project?

I think that would be a good idea. While I understand Marc's frustration
with these silly things, I also understand Jon and Peters position that
it really doesn't care what *we* think. The license isn't for us. The
license is for *other* people who want to help/use our code. And if
*they* interpret things another way, then that's what counts. Sad but
true.

A license is a set of words. Words are just to communicate. If the one
being communicated with do not understand, the set of words used to
communicate are wrong, no matter how right they are.

/Rickard




Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/28/2000 5:22 PM, "Peter Donald" <do...@locus.apache.org> wrote:

> Once RMS finds out
> about the project misusing the GPL he will start advocating all the GNU
> peopls stay away from it.

Someone want to send an email to licensing@gnu.org recommending that RMS
take a look at how the GPL is being used within the JBoss project?

:-)

-jon

-- 
http://scarab.tigris.org/    | http://noodle.tigris.org/
http://java.apache.org/      | http://java.apache.org/turbine/
http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
http://www.collab.net/       | http://www.sourcexchange.com/



Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Peter Donald <do...@locus.apache.org>.
At 02:18  28/10/00 -0700, you wrote:
>on 10/28/2000 10:05 AM, "marc fleury" <ma...@telkel.com> wrote:
>
>> I am sorry, I should actually provide some information.
>> 
>> We use the GPL to protect the kernel.  The virality of the GPL applies to
>> the "derived work" or "modified work as a whole" of the kernel.
>
>That is how you interpret it, not how RMS interprets it.

nor how software lawyers interpret it. While it has only been put to the
test a few times and they usually end up in out of court settlement all the
lawyers who have looked at it (both under pay of FSF and otherwise) have
not upheld this view. 

In very few circumstances is anything in the same JVM allowed to be non-GPL
unless it is covered by 3rd clause of GPL.

>The only solution for you is to choose a NON GPL license for JBoss such as
>MPL, BSD or APL. Period. You have no other choice unless you want to
>completely loose control of JBoss and have all of your hard excellent work
>completely ignored.

Thats one of the pities of GPL - it protects the code not the people ;(

>I'm not saying that I'm going to do it (I'm definitely not going to do it
>cause I don't like EJB anyway), but while I was at ApacheCon, I have
>certainly heard enough other people talking about doing it.

There is a few other oss solutions out there. I know of two groups who were
put off by jBoss'es license and searched for other OSS ejb servers. Only a
few are about but 

>People are VERY unhappy with the fact that JBoss is GPL and will be looking
>for other solutions to support.

now thats an understatement ;)

And no one has even invoked the wrath of RMS yet ;) Once RMS finds out
about the project misusing the GPL he will start advocating all the GNU
peopls stay away from it. His opinions while sometimes a little kafuie are
generally listened to. See some of the other projects he has attacked and
you will see a list of projects that have lost a lot of mind share compared
to what they should have (ie KDE, TCL are 2 good examples). Or even worse -
you may get slashdotted ;)

Cheers,

Pete

*------------------------------------------------------*
| "Nearly all men can stand adversity, but if you want |
| to test a man's character, give him power."          |
|       -Abraham Lincoln                               |
*------------------------------------------------------*

RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by marc fleury <ma...@telkel.com>.

|-----Original Message-----
|From: jboss-dev@list.working-dogs.com
|[mailto:jboss-dev@list.working-dogs.com]On Behalf Of Aaron Mulder
|Sent: Saturday, October 28, 2000 12:49 PM
|To: jBoss Developer; Java Apache Framework; Tomcat Dev List
|Subject: RE: [jBoss-Dev] Re: jboss on tomcat update
|
|
|Marc,
|	I'm on your side.  I would *really* like to be able to use jBoss.
|However:
|
|	Please interpret the following passage of the GPL, and tell us how
|it applies to an app server based on software like Apache, Tomcat, Avalon,
|Castor, Tyrex, PostgreSQL, and jBoss.  Specifically, imagine I want to
|make available a single download for a full open-source J2EE app server.
|Do you think it is possible to include jBoss with non-GPL packages?
|
|2.b.) You must cause any work that you distribute or publish, that in
|whole or in part contains or is derived from the Program or any part
|thereof, to be licensed as a whole at no charge to all third parties under
|the terms of this License.

Ok...
So a work that contains "part or whole" of another work, is one that
contains (maybe thinking import will help, or even cut and paste)  the work
to create the new work.  If you write a piece that "contains" my work (as in
import or cut and paste) then you must GPL, since the work contains part of
my work, btw that is the strongest case of containment (physical presence of
my work in your work), and of course GPL applies, "modified" or "derived"
are other level (less strong, you can "derive" from my work and not
physically "contain" it).

Now a work called "server" is **aggregation** and the pieces are not
"containment" nor "derived" nor "modified" of each other, they are
aggregated.  What is "derived" and "contained" in our case are the MBeans
adapters (they include a Logger) and these are GPL.  But Tomcat doesn't
contain or modify or derive from jboss.


Does a "webserver" contain linux or not?  is the work "apache" derived from
Linux, (some could say yes through system calls but "applications normal use
of sys lib" are not covered by Linus) does the work "apache" contain "part
or whole" of the work Linux.  ? .. right :)))

No! That is called "aggregation" and is explicitly not covered by the
license (read that passage).

YES it is possible to aggregate Apache+Tomcat (but isn't tomcat going to do
http as well as?;-), Avalon (F we need JMX man, come on the 77 board ;-)
Castor, Tyrex (cause assaf still kicks ass), PostgreSQL and jBOSS.

And yes!! it is possible to include GPL packages with non-GPL packages that
are not work containing the GPL work (any Linux distro will do ...)

marc


|
|Thanks,
|	Aaron
|
|On Sat, 28 Oct 2000, marc fleury wrote:
|> Ok,
|>
|> I am sorry, I should actually provide some information.
|>
|> We use the GPL to protect the kernel.  The virality of the GPL applies to
|> the "derived work" or "modified work as a whole" of the kernel.
|>
|> Tomcat is not "derived work" of jboss, clearly, wouldn't you say? :). The
|> "modified work as a whole" done in jboss to integrate the Tomcat
|jar is the
|> MBean adapter (for JMX), the Tomcat Interceptors (classLoaders), and the
|> J2EE deployer that we have developed.  Those are GPL, as per the
|GPL derived
|> work virality.
|>
|> The GPL applies to derived work in distribution.  Our
|distributions are GPL
|> kosher.
|> Please don't be afraid of it, and feel free to discuss it...
|>
|> regards
|>
|> marc
|>
|> |-----Original Message-----
|> |From: marc fleury [mailto:marc.fleury@telkel.com]
|> |Sent: Friday, October 27, 2000 10:10 PM
|> |To: jBoss Developer; tomcat-dev@jakarta.apache.org;
|> |java-apache-framework@list.working-dogs.com
|> |Subject: RE: [jBoss-Dev] Re: jboss on tomcat update
|> |
|> |
|> || but at the same time, you have a problem with the GPL being
|> ||viral so you give exceptions for people to use JBoss. Instead, what you
|> ||should do is probably be using the MPL license which will
|solve your needs
|> ||without having to constantly grant exceptions to people.
|> |
|> |???
|> |
|> |what 'exceptions'? we never granted 'exceptions'.  Please explain.
|> |
|> ||It is funny to me how you say that you are integrating our code which I
|> ||think is great, but the real issue is that we can't integrate YOUR code
|> ||because you choose to use the GPL license.
|> |
|> |why not?  what exactly prevents you from integrating our work?
|> |Please be explicit,
|> |
|> |let's not work from hearsay and "impressions" of the GPL, the GPL
|> |is very explicit.
|> |
|> |regards
|> |
|> |marc
|> |
|> |
|> |
|> ||
|> ||Sigh.
|> |
|> |????
|> |
|> ||
|> ||-jon
|> ||
|> ||
|> ||
|>
|>
|> ---------------------------------------------------------------------
|> To unsubscribe, e-mail: tomcat-dev-unsubscribe@jakarta.apache.org
|> For additional commands, e-mail: tomcat-dev-help@jakarta.apache.org
|>
|
|
|
|


RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by Aaron Mulder <am...@alumni.princeton.edu>.
Marc,
	I'm on your side.  I would *really* like to be able to use jBoss.  
However:

	Please interpret the following passage of the GPL, and tell us how
it applies to an app server based on software like Apache, Tomcat, Avalon,
Castor, Tyrex, PostgreSQL, and jBoss.  Specifically, imagine I want to
make available a single download for a full open-source J2EE app server.  
Do you think it is possible to include jBoss with non-GPL packages?

2.b.) You must cause any work that you distribute or publish, that in
whole or in part contains or is derived from the Program or any part
thereof, to be licensed as a whole at no charge to all third parties under
the terms of this License.

Thanks,
	Aaron

On Sat, 28 Oct 2000, marc fleury wrote:
> Ok,
> 
> I am sorry, I should actually provide some information.
> 
> We use the GPL to protect the kernel.  The virality of the GPL applies to
> the "derived work" or "modified work as a whole" of the kernel.
> 
> Tomcat is not "derived work" of jboss, clearly, wouldn't you say? :). The
> "modified work as a whole" done in jboss to integrate the Tomcat jar is the
> MBean adapter (for JMX), the Tomcat Interceptors (classLoaders), and the
> J2EE deployer that we have developed.  Those are GPL, as per the GPL derived
> work virality.
> 
> The GPL applies to derived work in distribution.  Our distributions are GPL
> kosher.
> Please don't be afraid of it, and feel free to discuss it...
> 
> regards
> 
> marc
> 
> |-----Original Message-----
> |From: marc fleury [mailto:marc.fleury@telkel.com]
> |Sent: Friday, October 27, 2000 10:10 PM
> |To: jBoss Developer; tomcat-dev@jakarta.apache.org;
> |java-apache-framework@list.working-dogs.com
> |Subject: RE: [jBoss-Dev] Re: jboss on tomcat update
> |
> |
> || but at the same time, you have a problem with the GPL being
> ||viral so you give exceptions for people to use JBoss. Instead, what you
> ||should do is probably be using the MPL license which will solve your needs
> ||without having to constantly grant exceptions to people.
> |
> |???
> |
> |what 'exceptions'? we never granted 'exceptions'.  Please explain.
> |
> ||It is funny to me how you say that you are integrating our code which I
> ||think is great, but the real issue is that we can't integrate YOUR code
> ||because you choose to use the GPL license.
> |
> |why not?  what exactly prevents you from integrating our work?
> |Please be explicit,
> |
> |let's not work from hearsay and "impressions" of the GPL, the GPL
> |is very explicit.
> |
> |regards
> |
> |marc
> |
> |
> |
> ||
> ||Sigh.
> |
> |????
> |
> ||
> ||-jon
> ||
> ||
> ||
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tomcat-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tomcat-dev-help@jakarta.apache.org
> 



[NOISE] Licenses (was: Re: [jBoss-Dev] Re: jboss on tomcat update)

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/28/2000 4:46 PM, "marc fleury" <ma...@telkel.com> wrote:

> |That is how you interpret it, not how RMS interprets it.
> 
> I have a license and the wording is clear.
> What people say he said isn't the question.
> 
> |I cannot take Tomcat and combine it with JBoss and make a
> |distribution of it
> |that is available from Apache.org because JBoss is under a GPL license.
> |Period.
> 
> again that is an ASF decision (clean tree policy), not a license problem per
> se.

Huh? The clean tree policy is completely related to the license problems of
the GPL.

> |Marc, I'm definitely not afraid to tell you what I think and I'm tired of
> |discussing it. I'm burnt out on this war. It is completely stupid and it
> |looks like you are going to have to get yourself burnt before you see the
> |light. Sigh.
> 
> wow, religious overtones...  you are going to "burn me to see the light".
> 
> You and what army, Torquemada?

I'm not going to burn you. I didn't say that at all. I leave my quote above
so that you can re-read it again and maybe understand it the second time you
read it.

> |The only solution for you is to choose a NON GPL license for JBoss such as
> |MPL, BSD or APL. Period. You have no other choice unless you want to
> |completely loose control of JBoss and have all of your hard excellent work
> |completely ignored.
> 
> the *only* final solution to what problem?
> the fact we can't live in your tree? well heck 60% of the world's OSS code
> lives in GPL...

Again, you repeat this 60% value that has absolutely no context or any proof
to back it up. Wake up, not everything is GPL.

> we already integrate, repeat, we already integrate!!!!

No, you integrate with us because our license allows you to. We cannot
integrate with you. That is where you don't quite get it.

> |I just went through this same exact stupid war with Justin Wells for nearly
> |4 months over WebMacro and he ended up releasing WM under the APL license
> |after he saw us take away his control of his product by producing a far
> |superior product under a APL license.
> 
> hee hee threats...
> And you wonder why we are happy doing things our way...

Not a threat at all. It is a factual statement of what happened and I'm
warning you that you will probably have something similar happen to your
product.

Good luck.

-jon


RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by marc fleury <ma...@telkel.com>.
|That is how you interpret it, not how RMS interprets it.

I have a license and the wording is clear.
What people say he said isn't the question.

|I cannot take Tomcat and combine it with JBoss and make a
|distribution of it
|that is available from Apache.org because JBoss is under a GPL license.
|Period.

again that is an ASF decision (clean tree policy), not a license problem per
se.

"period"

|Marc, I'm definitely not afraid to tell you what I think and I'm tired of
|discussing it. I'm burnt out on this war. It is completely stupid and it
|looks like you are going to have to get yourself burnt before you see the
|light. Sigh.

wow, religious overtones...  you are going to "burn me to see the light".

You and what army, Torquemada?

|The only solution for you is to choose a NON GPL license for JBoss such as
|MPL, BSD or APL. Period. You have no other choice unless you want to
|completely loose control of JBoss and have all of your hard excellent work
|completely ignored.

the *only* final solution to what problem?
the fact we can't live in your tree? well heck 60% of the world's OSS code
lives in GPL...

we already integrate, repeat, we already integrate!!!!

wake up! you are delirious! wake up!

|I just went through this same exact stupid war with Justin Wells for nearly
|4 months over WebMacro and he ended up releasing WM under the APL license
|after he saw us take away his control of his product by producing a far
|superior product under a APL license.

hee hee threats...
And you wonder why we are happy doing things our way...

|I'm not saying that I'm going to do it (I'm definitely not going to do it
|cause I don't like EJB anyway), but while I was at ApacheCon, I have
|certainly heard enough other people talking about doing it.
|
|People are VERY unhappy with the fact that JBoss is GPL and will be looking
|for other solutions to support.
|
|You have been warned.

Jon, this threatening is frankly unlike you.  There as always been oh...  35
competing projects in my sphere?... Now before you piss anyone off go take a
bath, sit back and relax, it's going to be fine, you are going to be
assimilated and all is going to be good...   I am not going to take it
seriously, you are obviously jetlagged or something (the burning light and
stuff isn't like you),


marc

"space to ladies, go easy on the tea, repeat..."


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/28/2000 10:05 AM, "marc fleury" <ma...@telkel.com> wrote:

> I am sorry, I should actually provide some information.
> 
> We use the GPL to protect the kernel.  The virality of the GPL applies to
> the "derived work" or "modified work as a whole" of the kernel.

That is how you interpret it, not how RMS interprets it.

> Tomcat is not "derived work" of jboss, clearly, wouldn't you say? :). The
> "modified work as a whole" done in jboss to integrate the Tomcat jar is the
> MBean adapter (for JMX), the Tomcat Interceptors (classLoaders), and the
> J2EE deployer that we have developed.  Those are GPL, as per the GPL derived
> work virality.

I cannot take Tomcat and combine it with JBoss and make a distribution of it
that is available from Apache.org because JBoss is under a GPL license.
Period.

> The GPL applies to derived work in distribution.  Our distributions are GPL
> kosher.
> Please don't be afraid of it, and feel free to discuss it...

Marc, I'm definitely not afraid to tell you what I think and I'm tired of
discussing it. I'm burnt out on this war. It is completely stupid and it
looks like you are going to have to get yourself burnt before you see the
light. Sigh.

The only solution for you is to choose a NON GPL license for JBoss such as
MPL, BSD or APL. Period. You have no other choice unless you want to
completely loose control of JBoss and have all of your hard excellent work
completely ignored.

I just went through this same exact stupid war with Justin Wells for nearly
4 months over WebMacro and he ended up releasing WM under the APL license
after he saw us take away his control of his product by producing a far
superior product under a APL license.

I'm not saying that I'm going to do it (I'm definitely not going to do it
cause I don't like EJB anyway), but while I was at ApacheCon, I have
certainly heard enough other people talking about doing it.

People are VERY unhappy with the fact that JBoss is GPL and will be looking
for other solutions to support.

You have been warned.

-jon

-- 
http://scarab.tigris.org/    | http://noodle.tigris.org/
http://java.apache.org/      | http://java.apache.org/turbine/
http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
http://www.collab.net/       | http://www.sourcexchange.com/


RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by marc fleury <ma...@telkel.com>.
Ok,

I am sorry, I should actually provide some information.

We use the GPL to protect the kernel.  The virality of the GPL applies to
the "derived work" or "modified work as a whole" of the kernel.

Tomcat is not "derived work" of jboss, clearly, wouldn't you say? :). The
"modified work as a whole" done in jboss to integrate the Tomcat jar is the
MBean adapter (for JMX), the Tomcat Interceptors (classLoaders), and the
J2EE deployer that we have developed.  Those are GPL, as per the GPL derived
work virality.

The GPL applies to derived work in distribution.  Our distributions are GPL
kosher.
Please don't be afraid of it, and feel free to discuss it...

regards

marc

|-----Original Message-----
|From: marc fleury [mailto:marc.fleury@telkel.com]
|Sent: Friday, October 27, 2000 10:10 PM
|To: jBoss Developer; tomcat-dev@jakarta.apache.org;
|java-apache-framework@list.working-dogs.com
|Subject: RE: [jBoss-Dev] Re: jboss on tomcat update
|
|
|| but at the same time, you have a problem with the GPL being
||viral so you give exceptions for people to use JBoss. Instead, what you
||should do is probably be using the MPL license which will solve your needs
||without having to constantly grant exceptions to people.
|
|???
|
|what 'exceptions'? we never granted 'exceptions'.  Please explain.
|
||It is funny to me how you say that you are integrating our code which I
||think is great, but the real issue is that we can't integrate YOUR code
||because you choose to use the GPL license.
|
|why not?  what exactly prevents you from integrating our work?
|Please be explicit,
|
|let's not work from hearsay and "impressions" of the GPL, the GPL
|is very explicit.
|
|regards
|
|marc
|
|
|
||
||Sigh.
|
|????
|
||
||-jon
||
||
||


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Peter Donald <do...@locus.apache.org>.
At 04:35  28/10/00 -0700, you wrote:
>on 10/28/2000 4:06 PM, "marc fleury" <ma...@telkel.com> wrote:
>
>> Indeed if the Avalon guy puts jBoss code in his tree and "contains" our
work
>> in his work then yeah.. that needs to be GPL.
>
>Bingo. So, this is something that is a major problem for me.

and me - I am that guy ;)

>> This is the mutations I was talking about... but the ASF decides not
>> to mingle with 60% of the world's OSS codebase *deliberately*.
>
>I'm sorry, but where exactly do you get that 60% number from?

60% is complete and utter BS with respect to java.  Perhaps in c or c++ ma
not with java.

>In my mind, OSS is about simply getting credit for the work that you do, not
>requiring people to either jump through hoops or give back their additions
>or changes to my source code. I don't give a rats ass what people do with my
>source code as long as they simply give me credit for my hard work.

right. Thats the difference one philosophical difference between GPL and
APL. GPL protects the code while APL protects the people. This results in a
number of different things - one of which is frequency of forking. APL
tends (from what I can see) to fork less often thou this may have a bit to
do with institution (especially 2 codebases in one project approach)

Cheers,

Pete

*------------------------------------------------------*
| "Nearly all men can stand adversity, but if you want |
| to test a man's character, give him power."          |
|       -Abraham Lincoln                               |
*------------------------------------------------------*

Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Peter Donald <do...@locus.apache.org>.
>	The problem is, I'm in a situation where (to quote "Ronin"),
>"Whenever there's a doubt, there is no doubt."  Whatever you say, I
>haven't heard anything that convinces me that the interpretation is clear
>- I can easily see both sides of the disagreement.  I suspect the only way
>for this to ultimately be resolved is to take it to a lawyer and/or RMS.  
>Any volunteers?  :)

I would strongly suggest you not mention the project by name and just use
hypotheticals. RMS has discussed this kind of situation before and you can
find all the information on his opinion regarding it in the philosophy
section of gnu. It may be best to just read through relevemt portions of
that instead of invoking his wrath ;)



Cheers,

Pete

*------------------------------------------------------*
| "Nearly all men can stand adversity, but if you want |
| to test a man's character, give him power."          |
|       -Abraham Lincoln                               |
*------------------------------------------------------*

Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Peter Donald <do...@locus.apache.org>.
>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.

power of choice is power of responsibility. You can not choose to claim to
distribute under X and violate X which is what I am irritated at.

>(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.
>;-)

or you have outsiders like me who use neither product ;)

>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.

right.

>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."

right.

>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.

right.

>So I guess that the question boils down to: When do we have "mere
>aggregation" and when do we have "combined software" ?

if it imports class/package, directly hardwires classnames, hardwires class
interfaces or hardwire other aspects it is called combined software. jBoss
does this with JMX, Tomcat and other non-GPL compatable libraries. This is
only allowed if it falls under clause 3 of GPL (neither of the above do) or
in other "special" circumstances* (neither of which do either).

* Special circumstances can basically be thought of as hosted components
(ie you could have EJBs in container without GPLing EJBs).

If packages are included in same (like the jars in bin/ and client/) then
they are combined software of which jBoss has a bit. jBoss violates the GPL
as this software is not GPL.

>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.

right.

>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.

APL license has nothing to do with it - it is the GPL license that is the
one that makes the restriction. And this would not brake the GPL because it
is a distribution of distributions.

>But is this ISO image "mere aggregation", or is it "combined
>software" ?

aggregation.

>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.

false.

>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.

They have been combine because jBoss hardwires interaction to tomcat jars.

>How do you think these two conflicting arguments would hold
>in a court of law?

first one falls down as does second thou IANAL and a lawyer should be
consulted.

>> 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.

nope.

>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 jBoss removes the tomcat linking classes, all MBeans, and the libraries
(and any other non-GPL compatable stuff) it would not violate the GPL but
it would also not be very useful now would it ?

>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.

again - nope.

>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?

yep - as long as it is a distribution of distributions then it is OK. Thou
jBoss still violates it with respect to all the stuff I said above.

>> 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.

right.

>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?

Not about the APL - the APL saids almost anything goes. It is the GPL that
is the license that causes the restriction. Defaulting to Tomcat classes is
fine as long as it is practicably possible to implement multiple webservers
with ease - which means not using any special features of tomcat and having
a "clean" interface.

>> >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.

too late - I got tired of discussing it and as jBoss is a serious threat to
GNU and they can eliminate the threat (just disallow jboss.org from using
license or derivatives until complied) he has a right to know.

I may work totally within APL now but I still respect the GPL - something
jBoss does not do. If push comes to shove it will most likely end up with
jBoss isolated from much of free software community. This will be a pity
but you have had enough oportunities to stand down but chose to go forward
with little respect for those around you.

>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.

right.

><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>

Depends on how you phrase it. From my experience he will give you both his
opinion and a legally defensible position. The GPL is relatively water
tight as far as licenses go and it has stood the test od time. Very few
people who violate the GPL have gone to court and very few who have gone to
court (zero I believe) have chose not to settle out-of-court before
proceedings. 

You could argue that this is because of all the bad karma etc but in
reality it is because the GPL is very explicit and it takes a lot of time
to understand and is very water tight.

Whatever the case jBoss violates the license in too many cases for it to be
worthwhile going to court. I don't intend to use it in particular so I have
no real problem if jBoss decides to choose a license that prohibts
cooperation. I think it is unfortunate and silly and would fully expect to
see it fall by the wayside but I really have no problem with it.

Whatever the case you have a window of about 3 days before RMS fires up and
possibly a week or two before it will hit the relevent sites. If in that
time you don't stop violation of GPL then you will feel the wrath of GNU
community (which includes those loons at slashdot). This will not  be
pleasant for any involved. I would urge you to try to either fcomply with
GPL or choose a new license in this time.



Cheers,

Pete

*------------------------------------------------------*
| "Nearly all men can stand adversity, but if you want |
| to test a man's character, give him power."          |
|       -Abraham Lincoln                               |
*------------------------------------------------------*

Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Rickard Öberg <ri...@telkel.com>.
Hi!

Jon Stevens wrote:
> > (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.
> > ;-)
> 
> Maybe so. Maybe that also says something about EJB. Maybe it is only
> something that should be provided by large corporations (ie: IBM/BEA) for
> exorbitant prices because it is something that is only really useful for
> such situations.

Hm.. you should have "IMHO"'ed that, because I find EJB very useful for
most things I do. I don't agree at all with these things being useful in
"only such situations".

In any case it's irrelevant to the discussion, so please leave it out.

/Rickard

-- 
Rickard Öberg

Email: rickard@telkel.com
http://www.telkel.com
http://www.jboss.org
http://www.dreambean.com


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/29/2000 11:19 PM, "Ole Husgaard" <os...@sparre.dk> wrote:

> 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.

Right, but at the core of the discussion IS the license so there is nothing
that you can do about it.

> 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.

It is larger than just the copyright because it is about Marc wanting to
protect his source code from the big bad corporations that are beating down
his door trying to steal his source code from him.

> (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.
> ;-)

Maybe so. Maybe that also says something about EJB. Maybe it is only
something that should be provided by large corporations (ie: IBM/BEA) for
exorbitant prices because it is something that is only really useful for
such situations.

-jon

-- 
http://scarab.tigris.org/    | http://noodle.tigris.org/
http://java.apache.org/      | http://java.apache.org/turbine/
http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
http://www.collab.net/       | http://www.sourcexchange.com/


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Ole Husgaard <os...@sparre.dk>.
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.

RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by Peter Donald <do...@locus.apache.org>.
>|	What can I say?  I agree that this is a reasonable interpretation.
>|But I don't think it's the only interpretation, and I'm not sure it's even
>|the interpretation intended by the authors.  There's another section that
>|specifically allows distribution of GPL and non-GPL programs on the same
>|medium (Linux distributions), and that passage would be redundant if this
>|passage reads as you suggest.
>
>Listen it says
>
>if work is "containing, modifying, deriving" (CMD) of work that is GPL then
>GPL.  If not, then not.
>
>(its' a mathematical if and only if)
>Ok let's loop on that for a while...
>
>Apache+Linux=aggregation, Apache is not CMD of Linux
>
>Frankly the wording is extremelly clear.  GPL applies to "contained",
>"modified", or "derived" work not aggregated work and that is in the
>license....
>
>what is not clear about it?

well obviously enough that you don't understand it. You really should seek
legal advice or at least advice from someone who knows what they are
talking about (RMS for the GNU angle).

>|	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.

>| The problem is, I'm in a situation where (to quote "Ronin"),
>|"Whenever there's a doubt, there is no doubt."  Whatever you say, I
>
>Listen, I like the ronin trick
>but there is NO doubt,
>
>Repeat if and only if work is CMD of GPL Work then GPL.

umm NO - go reread it again. Pay particular attention to all sections and
then explain why clause 3 has a number of exceptions near the end ? If you
are still not convinved then seek legal advice

>Take Tomcat, is that "containing" "deriving" "modifying" work of jboss?
>NOOOOO!!!!  you don't derive (although you might have been inspired by the
>interceptor layout;-) you don't modify (afaik) and you CERTAINLY don't
>contain.

if tomcat is contained in same archive then it has to be GPL. 
If code is contained in archive that links against tomcat then that code
has to be GPL. If that code is GPL then tomcat has to be GPL.

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)

This is a PITA yes and that is the intended solution. RMS set the license
up this way so that it makes it more desirable to GPL the whole work.

>|haven't heard anything that convinces me that the interpretation is clear
>|- I can easily see both sides of the disagreement.  I suspect the only way
>|for this to ultimately be resolved is to take it to a lawyer and/or RMS.
>|Any volunteers?  :)
>
>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 ?

>2- this is copyright telkel+jboss authors.

this has nothing to do with anything.

>|	Overall, the most unfortunate thing here is that I don't believe
>|either party is trying to lock out code from the other.  But the fact that
>|the licenses are not compatible means that one group or the other has to
>
>again the licenses are compatible, what is not compatible is that ASF
>doesn't want GPL code in the tree.  

NO they are NOT. RMS goes over the reasons on the web page. There is
currently one remaining clause that makes APL not GPL compatable (thou
hopefully eventually that will go away)

>It's a policy gentlemen, that forces us
>to do something if we want you guys to include code in your tree.  Again I
>understand that, well I respect it more than I understand it, and that seems
>to mean "separate projects".  

It is not the APL that blocks it but the GPL. GPL is incompatable with APL
and linking against it like you have done is illegal and all copyright
holders are liable for such things. 

I would hate to be in that position when a large company adopts jBoss
looses cash because you have mislicensed product and then goes after you.
In my country (Australia) it means up to 10 years in jail - sound like fun ?

>I already stated my honest belief that many
>bees are better than a drunken duck (wow, that sounded good, I gotta reuse
>it and use that ronin trick )...

I agree

>|change licenses in order to enable true sharing of code, which is one of
>|the greatest promises of open source.  And it doesn't sound like either
>|party is willing.
>
>to be very honest, I don't care that much, about the GPL, the APL, and the
>horses they rode in town on.  I care even less about the grandiose
>statements about "philosophy" from "the ladies corner".  Honestly all I care
>about is that our common code kicks ass.

Well considering you are butchering the GPL it is quite obvious you have no
respect for it. Whether you like it or not you have insulted a lot of
people by this - maybe it will stand and maybe it won't but consider this
way. The GNU projects sees projects like your own who violate the GPL as
bigger threats than proprietry software, Apache considers them not useful
as they can't share. Put this in a bowl and shake and you have your
situation. 


Cheers,

Pete

*------------------------------------------------------*
| "Nearly all men can stand adversity, but if you want |
| to test a man's character, give him power."          |
|       -Abraham Lincoln                               |
*------------------------------------------------------*

RE: [BUG TRACKING] Searches disabled

Posted by "Rob S." <rs...@home.com>.
Oops, apologies.  Meant that for Nick only =)

- r

> -----Original Message-----
> From: Rob S. [mailto:rslifka@home.com]
> Sent: October 30, 2000 8:51 PM
> To: tomcat-dev@jakarta.apache.org
> Subject: RE: [BUG TRACKING] Searches disabled
>
>
> What's needed?
>
> - r
>
> > -----Original Message-----
> > From: Nick Bauman [mailto:nick@cortexity.com]
> > Sent: October 30, 2000 8:48 PM
> > To: tomcat-dev
> > Subject: [BUG TRACKING] Searches disabled
> >
> >
> > I've disabled searches on Jakarta BugRat. You will just have to browse
> > the bugs until I untangle to code that does the searches. Suffice it to
> > say I can't imagine a more resource-intensive series of queries than the
> > ones that are used for bug and report searching in BugRat. Volunteers to
> > help?
> >
> > -Nick
> >
> >
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: tomcat-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: tomcat-dev-help@jakarta.apache.org
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tomcat-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tomcat-dev-help@jakarta.apache.org
>


RE: [BUG TRACKING] Searches disabled

Posted by "Rob S." <rs...@home.com>.
What's needed?

- r

> -----Original Message-----
> From: Nick Bauman [mailto:nick@cortexity.com]
> Sent: October 30, 2000 8:48 PM
> To: tomcat-dev
> Subject: [BUG TRACKING] Searches disabled
> 
> 
> I've disabled searches on Jakarta BugRat. You will just have to browse
> the bugs until I untangle to code that does the searches. Suffice it to
> say I can't imagine a more resource-intensive series of queries than the
> ones that are used for bug and report searching in BugRat. Volunteers to
> help?
> 
> -Nick
> 
> 
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tomcat-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: tomcat-dev-help@jakarta.apache.org
> 

[BUG TRACKING] Searches disabled

Posted by Nick Bauman <ni...@cortexity.com>.
I've disabled searches on Jakarta BugRat. You will just have to browse
the bugs until I untangle to code that does the searches. Suffice it to
say I can't imagine a more resource-intensive series of queries than the
ones that are used for bug and report searching in BugRat. Volunteers to
help?

-Nick




Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Federico Barbieri <sc...@betaversion.org>.
> on 10/29/2000 7:57 PM, "Nick Bauman" <ni...@cortexity.com> wrote:
> 
> > Right: we aren't talking about fine wine, rare stamps or gold boullion
> > here. The code has to move and be moveable to live, to be of value.
> 

Exactly. Code is not important. People are. You may think a large
company can "steal" your code becouse they have more _coding_power_ that
you have but as I already told you I don't belive this can ever happend
becouse there is no company that has more _brain_power_ than an open
source comunity. 

> > Marc insists that GPL protects young code. I don't buy that either.

what does it means "GPL protect my code"? 
As you know I'd like to use jbos proxy generation code for avalon but I
can't for GPL is specifically designed to prevent this. Now why you
think this could damage jboss? IMNSHO such a code sharing can just
advantage both of us. I get good code and provide better features to
avalon (without forcing JDK1.3) and you get credits. Don't you like the
idea that everybody is using your code becouse it's good code? Isn't
this the best reward for a developer? 

You may think I'm a good guy so I will give credit but there are many
bad guys around like in corporate that can steal you code and not give
credit. 
But:
- APL is designed for this
- nobody can hide such a thing from the comunity
- if your code is really good company will be happy to say "our product
is based on jboss code!".

So again I don't see any good reason for keeping this code GPL but I can
see many many reason not to do that.

Federico 

fede@apache.org

Re: [NOISE] [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/29/2000 8:33 PM, "Nick Bauman" <ni...@cortexity.com> wrote:

> An aside,
> 
> There is, AFAIK, one good reason to use GPL over any other Open Source or
> Free Software license, and it's a very very good reason: To maximize the
> spread of the GPL.
> 
> IOW, it's to forward the tenets of freedom in software development and to
> more or less declare that other software is opposed to freedom. BSD-style
> licenses are trying to co-exist with closed software. GPL is trying to
> fight closed software. With GPL, it's a case of "the friend of my enemy is
> my enemy".
> 
> I personally like the GPL for this reason, but then I am a tree-hugging
> pinko, like RMS. Not many of my colleages are and Eric Raymond for damn
> sure is not. (he would probably shoot me picketing a sawmill with one
> of his handguns) But if I want to co-exist with other
> non-tree-hugging-non-pinkos I'd better learn the art of compromise. :)
> 
> -Nick

I live in Berkeley California with all those tree huggers. I like trees as
well. 

Reality is that I don't need a license to grow trees in the wild in order to
keep them free.

:-)

-jon

-- 
http://scarab.tigris.org/    | http://noodle.tigris.org/
http://java.apache.org/      | http://java.apache.org/turbine/
http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
http://www.collab.net/       | http://www.sourcexchange.com/



Re: [NOISE] [jBoss-Dev] Re: jboss on tomcat update

Posted by Nick Bauman <ni...@cortexity.com>.
An aside,

There is, AFAIK, one good reason to use GPL over any other Open Source or
Free Software license, and it's a very very good reason: To maximize the
spread of the GPL. 

IOW, it's to forward the tenets of freedom in software development and to
more or less declare that other software is opposed to freedom. BSD-style
licenses are trying to co-exist with closed software. GPL is trying to
fight closed software. With GPL, it's a case of "the friend of my enemy is
my enemy".

I personally like the GPL for this reason, but then I am a tree-hugging
pinko, like RMS. Not many of my colleages are and Eric Raymond for damn
sure is not. (he would probably shoot me picketing a sawmill with one
of his handguns) But if I want to co-exist with other 
non-tree-hugging-non-pinkos I'd better learn the art of compromise. :)

-Nick

On Sun, 29 Oct 2000, Jon Stevens wrote:

> > Marc insists that GPL protects young code. I don't buy that either.
> 
> LOL! That makes me laugh. So, when JBoss grows up, it can switch to a
> license for mature code (ie: BSD). :-)
> 
> It sounds like JBoss is getting more mature. So, that argument suggests even
> further that there is no reason for JBoss to be GPL.
> 
> > If I GPL to protect my young code I assume my code is vulnerable to
> > someone putting it into a commercial product and selling it where I would
> > miss out on the revenue? Pshaw! There is no viable motive there. And every
> > line of code I subseqently write undermines their commercial position.
> 
> Bingo.
> 
> > Or another reason is because I'm afraid someone *gasp* will incorporate it
> > into their other BSD-style licensed project and steal my mindshare /
> > marketshare? So someone with more time and money and expertise
> > might do a better job? Isn't that what we want? Better code? Or is
> > _control_ over my crap-tastic code a better thing? I don't think so.
> 
> LOL!
> 
> -jon
> 
> 

-- 
Nicolaus Bauman
Software Engineer
Simplexity Systems



Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/29/2000 7:57 PM, "Nick Bauman" <ni...@cortexity.com> wrote:

> Right: we aren't talking about fine wine, rare stamps or gold boullion
> here. The code has to move and be moveable to live, to be of value.

I agree.

> Marc insists that GPL protects young code. I don't buy that either.

LOL! That makes me laugh. So, when JBoss grows up, it can switch to a
license for mature code (ie: BSD). :-)

It sounds like JBoss is getting more mature. So, that argument suggests even
further that there is no reason for JBoss to be GPL.

> If I GPL to protect my young code I assume my code is vulnerable to
> someone putting it into a commercial product and selling it where I would
> miss out on the revenue? Pshaw! There is no viable motive there. And every
> line of code I subseqently write undermines their commercial position.

Bingo.

> Or another reason is because I'm afraid someone *gasp* will incorporate it
> into their other BSD-style licensed project and steal my mindshare /
> marketshare? So someone with more time and money and expertise
> might do a better job? Isn't that what we want? Better code? Or is
> _control_ over my crap-tastic code a better thing? I don't think so.

LOL!

-jon

-- 
http://scarab.tigris.org/    | http://noodle.tigris.org/
http://java.apache.org/      | http://java.apache.org/turbine/
http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
http://www.collab.net/       | http://www.sourcexchange.com/


RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by Peter Donald <do...@locus.apache.org>.
At 01:39  29/10/00 -0800, marc fleury wrote:
...some truly misguided stuff...

Oh - so I take that as "No I haven't contacted lawyers nor anyone who knows
what they are talking about". Well considering you have been made aware on
a publically archived list you really will have no defense in court if it
ever comes to such a case.

Just one question thou. Do you think that GPL code can import non-GPL
compatable code (that is not covered by clause 3) ? 

Cheers,

Pete

*------------------------------------------------------*
| "Nearly all men can stand adversity, but if you want |
| to test a man's character, give him power."          |
|       -Abraham Lincoln                               |
*------------------------------------------------------*

RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by Nick Bauman <ni...@cortexity.com>.
On Sun, 29 Oct 2000, marc fleury wrote:

> 
> THIS IS WHERE THE GPL DRAWS THE LINE FOR VIRALITY
> 
> 4 Aggregation is the weakest, it just means bundling of work.  GPL doesn't
> apply.
> 

Which to me means that the closest together the two can ever be is if
Tomcat talks to JBoss and vice versa via a network socket. Then the two
licenses can co-exist. Any code written to accept a Java interface after
that network socket speaks would negate the legality, so you are stuck
with something like http as your protocol. So why not just resort to
sharp sticks and rocks while we're at it?

But then as someone just mentioned, it matters not a stitch what you or I
or Jon says, it matter what they lawyers say.

-- 
Nicolaus Bauman
Software Engineer
Simplexity Systems



RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by marc fleury <ma...@telkel.com>.
ok jon et al...

it is trivial that APL software is incompatible with GPL, since for APL
software to "contain, derive or modify" GPL software is not possible, we all
agree on that.  Aggregation of it is OK per the GPL.

The point was and is on ordering of containment that I think is creating
confusion

++++++++++++++++++++++++++++++++++++++++++++++++++

Containment>>modification>>derivation>>aggregation

++++++++++++++++++++++++++++++++++++++++++++++++++

1 Containment is very strong, it means an import, a code copy paste, to
create your code.

2 Modification is strong (less) it just means not an exact import but you
modify the initial code and include it in yours.

3 Derivation is a weak link, it means "inspired by" starting from and
deriving code, a good example of this is the
"MBeans that we use to integrate Tomcat in jboss" (actually we also
"contain" the logger ...).

The virality of the GPL applies to work growing from work covered in 1/2/3

THIS IS WHERE THE GPL DRAWS THE LINE FOR VIRALITY

4 Aggregation is the weakest, it just means bundling of work.  GPL doesn't
apply.


As you know (and say) the GPL was devised to protect the authors of code by
protecting the code and the recipients of that code.  I.e. CMD of my work
WILL REMAIN GPL by viraling the CMD work.  CMD of jboss we will see, non CMD
of jboss, we can't talk to you... if someone aggregates tomcat with it, we
don't see it, the GPL doesn't require it, there is NOTHING we can do to
legally force you, your work is not CMD of ours... These are VERY different
things, very different levels of "containment" of work A to B.  And please
let's stop 10,000 feet philosophy!!! the wording is clear, legal.

jboss + tomcat (work as the server ) is aggregation mostly with some
"derivation" MBeans.
We all agree that not a line of tomcat falls under 1/2/3
Aggregation is clearly not covered by the license, and aggregating jboss and
tomcat is clearly not covered by GPL.

if Tomcat CONTAINS/MODIFIES/DERIVES "part or whole" of jboss (which it
doesn't) then GPL applies.  It is not interpretation it is extremelly clear.
The difference between Containment and aggregation that frankly seems to fly
over many heads here is introduced with many intermediary levels in the GPL.
Don't let "vapors of GPL" and "vague impressions" of what viral means, drive
your thinking (or thrive your drinking).  It is CLEARLY !!!! defined and
please take the time to understand the above...

So I am tired of repeating (and it seems each side is repeating at this
point so let's stop it) that Tomcat is not in any way related (meaning CMD),
so we would never require that you GPL your software for someone
distributing independent work in aggregation(not CMD).

BTW it would be us to require it not anybody else...  Also I will repeat and
draw your attention to the fact that this software is copyright by us not
FSF.... do you understand the implications of this, Mr Donald??? (and the
"10 years in Australia").  what anyone says is orthogonal, the license is by
us to you in the *wording* of the FSF's GPL, but it is not FSF's copyright,
so not in any way FSF's property, do you copy me?...

And frankly that is a "Standalone" CMD definition... it is extremelly clear.

again this is OUR (c), the last to leave this room please turn the light,
unless Mr Donald is still yelling "10 years in Australia, 10 years you
mo*her f&ckers!!!! Australia!!! 10 years!!!!" in a corner... (actually let's
just turn the light off and go)

Finally jon, I will agree with you on the "threats".  It's happened already,
and we are better for it.  I will draw your attention to the fact that the
EJB field is already full of alternatives, and the scenario you describe
already happened (someone took our work and duplicated it 9 month ago),
claiming "license" differences.   I recommend you look at JOnAS (using
enhydra as front end instead of tomcat or resin, can't blame them) and
openEJB (that's the one, just a container and an interesting direction, we
might integrate down the road).  Our field is very competitive, and again,
please leave the "apache" ego at the door, this is the java field and we
still have much to prove and APIs to get under our belts before we go around
like this... do you read me???


marc

________________________
"Don't do it be sincere"
-- Nude Dimensions 2--
________________________





|-----Original Message-----
|From: jboss-dev@list.working-dogs.com
|[mailto:jboss-dev@list.working-dogs.com]On Behalf Of Jon Stevens
|Sent: Sunday, October 29, 2000 11:00 AM
|To: tomcat-dev@jakarta.apache.org
|Cc: jboss-dev@list.working-dogs.com;
|java-apache-framework@list.working-dogs.com
|Subject: Re: [jBoss-Dev] Re: jboss on tomcat update
|
|
|on 10/29/2000 8:17 AM, "Nick Bauman" <ni...@cortexity.com> wrote:
|
|> Futhermore, I can change what I think is "contained" and what I think is
|> "aggregated" in a Java program in a very technical fashion. So far I
|> think Jon has a very salient point.
|
|Thanks. :-)
|
|Let me re-state things one more time though...the real "point" is that I
|spent 4 months trying to work on finding a way to work together
|with the GPL
|and the APL license with some of the "top" people in the OSS industry as
|well as trying to convince Justin Wells to either change to a MPL or APL
|license.
|
|Nothing worked because the GPL is simply unacceptable for many many people.
|Justin refused to do anything to change the license to something that the
|ASF would accept and he backed me into a corner where I had no other choice
|than to simply re-invent his software. Now, at this point, I'm very glad I
|did that because even though we may have forked the community for the short
|term, we have certainly come up with a far better product for the community
|for the long term as well as the fact that Velocity now has an ASF style
|community which I enjoy much more than a single dictator style community.
|
|Again, I'm not threatening that I will do this, but I'm warning Marc that
|someone else will definitely be doing this if he doesn't watch
|out. There is
|enough bad religious feelings about the GPL now and as far as I can tell,
|very few bad religious feelings about the BSD licenses. This in itself has
|already been proven to be enough cause to fork a project.
|
|Based on past experiences, I'm predicting the future and I'm just worried
|that Marc is going to continue arguing with people over this issue and
|someone will come along and simply take his hard work and duplicate it.
|Again, this stuff isn't rocket science. The rule so far is that it is far
|easier to write something from scratch than it is to deal with license
|issues.
|
|I know that is wrong, but the current religious license views that people
|have dictate that and I don't see it changing until RMS decides to change
|the GPL to allow people more freedom with what they do with other peoples
|source code.
|
|FYI, on this page (below), GNU specifically states that the APL 1.1 license
|is NOT compatible with the GPL license. So, Marc, whatever arguments that
|you are trying to make about being able to work with the APL and the GPL
|combined, you are wrong. It is clear that these two licenses cannot have
|their source code together in any shape or form without potentially risking
|the wrath of RMS or lawyers. It just isn't worth it to me to risk whatever
|project I am working on on trying to work with GPL'd software.
|
|<http://www.gnu.org/philosophy/license-list.html>
|
|thanks,
|
|-jon
|
|--
|http://scarab.tigris.org/    | http://noodle.tigris.org/
|http://java.apache.org/      | http://java.apache.org/turbine/
|http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
|http://www.collab.net/       | http://www.sourcexchange.com/
|
|
|
|


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/29/2000 8:17 AM, "Nick Bauman" <ni...@cortexity.com> wrote:

> Futhermore, I can change what I think is "contained" and what I think is
> "aggregated" in a Java program in a very technical fashion. So far I
> think Jon has a very salient point.

Thanks. :-)

Let me re-state things one more time though...the real "point" is that I
spent 4 months trying to work on finding a way to work together with the GPL
and the APL license with some of the "top" people in the OSS industry as
well as trying to convince Justin Wells to either change to a MPL or APL
license. 

Nothing worked because the GPL is simply unacceptable for many many people.
Justin refused to do anything to change the license to something that the
ASF would accept and he backed me into a corner where I had no other choice
than to simply re-invent his software. Now, at this point, I'm very glad I
did that because even though we may have forked the community for the short
term, we have certainly come up with a far better product for the community
for the long term as well as the fact that Velocity now has an ASF style
community which I enjoy much more than a single dictator style community.

Again, I'm not threatening that I will do this, but I'm warning Marc that
someone else will definitely be doing this if he doesn't watch out. There is
enough bad religious feelings about the GPL now and as far as I can tell,
very few bad religious feelings about the BSD licenses. This in itself has
already been proven to be enough cause to fork a project.

Based on past experiences, I'm predicting the future and I'm just worried
that Marc is going to continue arguing with people over this issue and
someone will come along and simply take his hard work and duplicate it.
Again, this stuff isn't rocket science. The rule so far is that it is far
easier to write something from scratch than it is to deal with license
issues. 

I know that is wrong, but the current religious license views that people
have dictate that and I don't see it changing until RMS decides to change
the GPL to allow people more freedom with what they do with other peoples
source code.

FYI, on this page (below), GNU specifically states that the APL 1.1 license
is NOT compatible with the GPL license. So, Marc, whatever arguments that
you are trying to make about being able to work with the APL and the GPL
combined, you are wrong. It is clear that these two licenses cannot have
their source code together in any shape or form without potentially risking
the wrath of RMS or lawyers. It just isn't worth it to me to risk whatever
project I am working on on trying to work with GPL'd software.

<http://www.gnu.org/philosophy/license-list.html>

thanks,

-jon

-- 
http://scarab.tigris.org/    | http://noodle.tigris.org/
http://java.apache.org/      | http://java.apache.org/turbine/
http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
http://www.collab.net/       | http://www.sourcexchange.com/



RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by Nick Bauman <ni...@cortexity.com>.
On Sat, 28 Oct 2000, marc fleury wrote:

> |	What can I say?  I agree that this is a reasonable interpretation.
> |But I don't think it's the only interpretation, and I'm not sure it's even
> |the interpretation intended by the authors.  There's another section that
> |specifically allows distribution of GPL and non-GPL programs on the same
> |medium (Linux distributions), and that passage would be redundant if this
> |passage reads as you suggest.
> 
> Listen it says
> 
> if work is "containing, modifying, deriving" (CMD) of work that is GPL then
> GPL.  If not, then not.
> 
> (its' a mathematical if and only if)
> Ok let's loop on that for a while...
> 
> Apache+Linux=aggregation, Apache is not CMD of Linux
> 
> Frankly the wording is extremelly clear.  GPL applies to "contained",
> "modified", or "derived" work not aggregated work and that is in the
> license....
> 
> what is not clear about it?
> 

I'll tell you, Marc, the word "contained" and the word "aggregated" as
far as Java software goes, is what is extreemely unclear. The OO gurus 
do not even agree what the difference is between "contained" and
"aggregated".

Futhermore, I can change what I think is "contained" and what I think is  
"aggregated" in a Java program in a very technical fashion. So far I 
think Jon has a very salient point.

-- 
Nicolaus Bauman
Software Engineer
Simplexity Systems



RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by marc fleury <ma...@telkel.com>.
|	This is truly unfortunate.  There are definitely ares of code that
|could be shared - that *should* be shared, such as logging, dynamic
|proxies, thread pools, and so on.  It's too bad that it doesn't happen
|until a javax package is available...  Particularly since those are *not*
|open source (well, under a much more restrictive license, anyway).

javax will solve the problem for the logging, the dynamic proxies are in 1.3
(if you want 1.2.2. then you need our stuff) the thread pools we might be
very interested in, but *we* can "contain" your code so :)))s.

Hey why don't you tell me *exactly* what you need, and we will see what we
can do to accomodate your "clean tree" policy.

|	What can I say?  I agree that this is a reasonable interpretation.
|But I don't think it's the only interpretation, and I'm not sure it's even
|the interpretation intended by the authors.  There's another section that
|specifically allows distribution of GPL and non-GPL programs on the same
|medium (Linux distributions), and that passage would be redundant if this
|passage reads as you suggest.

Listen it says

if work is "containing, modifying, deriving" (CMD) of work that is GPL then
GPL.  If not, then not.

(its' a mathematical if and only if)
Ok let's loop on that for a while...

Apache+Linux=aggregation, Apache is not CMD of Linux

Frankly the wording is extremelly clear.  GPL applies to "contained",
"modified", or "derived" work not aggregated work and that is in the
license....

what is not clear about it?

|	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

| The problem is, I'm in a situation where (to quote "Ronin"),
|"Whenever there's a doubt, there is no doubt."  Whatever you say, I

Listen, I like the ronin trick
but there is NO doubt,

Repeat if and only if work is CMD of GPL Work then GPL.

Take Tomcat, is that "containing" "deriving" "modifying" work of jboss?
NOOOOO!!!!  you don't derive (although you might have been inspired by the
interceptor layout;-) you don't modify (afaik) and you CERTAINLY don't
contain.


|haven't heard anything that convinces me that the interpretation is clear
|- I can easily see both sides of the disagreement.  I suspect the only way
|for this to ultimately be resolved is to take it to a lawyer and/or RMS.
|Any volunteers?  :)

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,
2- this is copyright telkel+jboss authors.


|	Overall, the most unfortunate thing here is that I don't believe
|either party is trying to lock out code from the other.  But the fact that
|the licenses are not compatible means that one group or the other has to

again the licenses are compatible, what is not compatible is that ASF
doesn't want GPL code in the tree.  It's a policy gentlemen, that forces us
to do something if we want you guys to include code in your tree.  Again I
understand that, well I respect it more than I understand it, and that seems
to mean "separate projects".  I already stated my honest belief that many
bees are better than a drunken duck (wow, that sounded good, I gotta reuse
it and use that ronin trick )...

|change licenses in order to enable true sharing of code, which is one of
|the greatest promises of open source.  And it doesn't sound like either
|party is willing.

to be very honest, I don't care that much, about the GPL, the APL, and the
horses they rode in town on.  I care even less about the grandiose
statements about "philosophy" from "the ladies corner".  Honestly all I care
about is that our common code kicks ass.

when all is said and all is done,  IT DOES.  Tomcat+jboss is a great j2ee
stack, a credible alternative to a billion$ market, welcome ladies, start
fighting cause the commercial EJB guys in front are not joking.

Ok, I need to go now, I have a costumed halloween party and I need to dress
up as Zorro...

<whistle/>
good night...

marc


|
|Aaron
|
|
|


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Nick Bauman <ni...@cortexity.com>.
On Sun, 29 Oct 2000, Jon Stevens wrote:

> on 10/29/2000 6:08 PM, "Nick Bauman" <ni...@cortexity.com> wrote:
> 
> > Which to me means that the closest together the two can ever be is if
> > Tomcat talks to JBoss and vice versa via a network socket. Then the two
> > licenses can co-exist. Any code written to accept a Java interface after
> > that network socket speaks would negate the legality, so you are stuck
> > with something like http as your protocol. So why not just resort to
> > sharp sticks and rocks while we're at it?
> > 
> > But then as someone just mentioned, it matters not a stitch what you or I
> > or Jon says, it matter what they lawyers say.
> 
> Exactly...why not just simplify things for everyone involved and make it a
> BSD license. It is the lowest common denominator that still provides
> protection for receiving credit for the work you do (that is all that I
> personally think OSS authors should get).

Right: we aren't talking about fine wine, rare stamps or gold boullion
here. The code has to move and be moveable to live, to be of value.
 
> So far, I haven't even seen one valid excuse for using the GPL for JBoss in
> the first place. Bluntly, this whole debate is simply around the fact that
> Marc doesn't want to own up to the fact that he choose a bad license (the
> GPL) for his software and doesn't want to admit that he was wrong after
> everyone (including myself) told him that the GPL was a bad decision.

Well now that's reading in a motive. Let's not back anyone into a 
corner: let's focus on a hypothetical common ground. Marc is trying to do
what he thinks is right. Jon has put forth some really good reasons why
it's not right. Marc has not really convinced me he's onto something in
his refutation of what Jon is saying. 

> The *really* silly thing here is that Marc thinks that by using the GPL he
> is protected against certain things when in reality the GPL doesn't protect
> him at all from what he wants protection for!

Marc insists that GPL protects young code. I don't buy that either. 

If I GPL to protect my young code I assume my code is vulnerable to
someone putting it into a commercial product and selling it where I would
miss out on the revenue? Pshaw! There is no viable motive there. And every
line of code I subseqently write undermines their commercial position.

Or another reason is because I'm afraid someone *gasp* will incorporate it
into their other BSD-style licensed project and steal my mindshare /
marketshare? So someone with more time and money and expertise
might do a better job? Isn't that what we want? Better code? Or is
_control_ over my crap-tastic code a better thing? I don't think so.

> 
> -jon
> 

Nick



Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/29/2000 6:08 PM, "Nick Bauman" <ni...@cortexity.com> wrote:

> Which to me means that the closest together the two can ever be is if
> Tomcat talks to JBoss and vice versa via a network socket. Then the two
> licenses can co-exist. Any code written to accept a Java interface after
> that network socket speaks would negate the legality, so you are stuck
> with something like http as your protocol. So why not just resort to
> sharp sticks and rocks while we're at it?
> 
> But then as someone just mentioned, it matters not a stitch what you or I
> or Jon says, it matter what they lawyers say.

Exactly...why not just simplify things for everyone involved and make it a
BSD license. It is the lowest common denominator that still provides
protection for receiving credit for the work you do (that is all that I
personally think OSS authors should get).

So far, I haven't even seen one valid excuse for using the GPL for JBoss in
the first place. Bluntly, this whole debate is simply around the fact that
Marc doesn't want to own up to the fact that he choose a bad license (the
GPL) for his software and doesn't want to admit that he was wrong after
everyone (including myself) told him that the GPL was a bad decision.

The *really* silly thing here is that Marc thinks that by using the GPL he
is protected against certain things when in reality the GPL doesn't protect
him at all from what he wants protection for!

All I can do at this point is sigh and shake my head in complete amazement.

-jon

-- 
http://scarab.tigris.org/    | http://noodle.tigris.org/
http://java.apache.org/      | http://java.apache.org/turbine/
http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
http://www.collab.net/       | http://www.sourcexchange.com/


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/28/2000 11:47 PM, "marc fleury" <ma...@telkel.com> wrote:

> jon,
> 
> True! it is not a "bad thing" and the APL is less restrictive.  GPL'ing the
> kernel makes sense for us as we fight an uphill battle.  We put good code
> and need exchange to grow... it is a "protect the young" kinda thing, the
> "feedback" is feeding the group. Important for a baby.
> 
> When the field is stabilized then it is another story... do you understand
> our position better?  Like I said let's work around the "clean tree policy"
> for things your really want in your tree from us.  For the rest it's clear
> on licenses in integration.

I'm sorry, I don't understand your position at all. I have started and lead
numerous large scale OSS projects, including Jserv and Tomcat without the
same fears that you seem to have for some unknown reason other than it
appears as though you don't understand how OSS works very well.

-jon

-- 
http://scarab.tigris.org/    | http://noodle.tigris.org/
http://java.apache.org/      | http://java.apache.org/turbine/
http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
http://www.collab.net/       | http://www.sourcexchange.com/



RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by marc fleury <ma...@telkel.com>.
|The amazing thing here is that the APL 1.1 license is one of the least
|restrictive licenses out there and definitely much less
|restrictive than the
|GPL. So, we are asking to not go to a MORE restrictive license, but to a
|LESS restrictive license. How can that be a bad thing?

jon,

True! it is not a "bad thing" and the APL is less restrictive.  GPL'ing the
kernel makes sense for us as we fight an uphill battle.  We put good code
and need exchange to grow... it is a "protect the young" kinda thing, the
"feedback" is feeding the group. Important for a baby.

When the field is stabilized then it is another story... do you understand
our position better?  Like I said let's work around the "clean tree policy"
for things your really want in your tree from us.  For the rest it's clear
on licenses in integration.

kind regards

marc

|
|-jon
|
|--
|http://scarab.tigris.org/    | http://noodle.tigris.org/
|http://java.apache.org/      | http://java.apache.org/turbine/
|http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
|http://www.collab.net/       | http://www.sourcexchange.com/
|
|
|
|


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/28/2000 5:41 PM, "Aaron Mulder" <am...@alumni.princeton.edu> wrote:

> Overall, the most unfortunate thing here is that I don't believe
> either party is trying to lock out code from the other.  But the fact that
> the licenses are not compatible means that one group or the other has to
> change licenses in order to enable true sharing of code, which is one of
> the greatest promises of open source.  And it doesn't sound like either
> party is willing.
> 
> Aaron

The amazing thing here is that the APL 1.1 license is one of the least
restrictive licenses out there and definitely much less restrictive than the
GPL. So, we are asking to not go to a MORE restrictive license, but to a
LESS restrictive license. How can that be a bad thing?

-jon

-- 
http://scarab.tigris.org/    | http://noodle.tigris.org/
http://java.apache.org/      | http://java.apache.org/turbine/
http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
http://www.collab.net/       | http://www.sourcexchange.com/



Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Aaron Mulder <am...@alumni.princeton.edu>.
On Sat, 28 Oct 2000, Jon Stevens wrote:
> on 10/28/2000 4:06 PM, "marc fleury" <ma...@telkel.com> wrote:
> > Indeed if the Avalon guy puts jBoss code in his tree and "contains" our work
> > in his work then yeah.. that needs to be GPL.
> Bingo. So, this is something that is a major problem for me.

	This is truly unfortunate.  There are definitely ares of code that
could be shared - that *should* be shared, such as logging, dynamic
proxies, thread pools, and so on.  It's too bad that it doesn't happen
until a javax package is available...  Particularly since those are *not*
open source (well, under a much more restrictive license, anyway).

On Sat, 28 Oct 2000, marc fleury wrote:
> |2.b.) You must cause any work that you distribute or publish, that in
> |whole or in part contains or is derived from the Program or any part
> |thereof, to be licensed as a whole at no charge to all third parties under
> |the terms of this License.
>
> again, "work that contains the Program" is code that "contains" physically 
> the Program (maybe thinking import in programming terms can help).

	What can I say?  I agree that this is a reasonable interpretation.  
But I don't think it's the only interpretation, and I'm not sure it's even
the interpretation intended by the authors.  There's another section that
specifically allows distribution of GPL and non-GPL programs on the same
medium (Linux distributions), and that passage would be redundant if this
passage reads as you suggest.
	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.
	The problem is, I'm in a situation where (to quote "Ronin"),
"Whenever there's a doubt, there is no doubt."  Whatever you say, I
haven't heard anything that convinces me that the interpretation is clear
- I can easily see both sides of the disagreement.  I suspect the only way
for this to ultimately be resolved is to take it to a lawyer and/or RMS.  
Any volunteers?  :)

	Overall, the most unfortunate thing here is that I don't believe
either party is trying to lock out code from the other.  But the fact that
the licenses are not compatible means that one group or the other has to
change licenses in order to enable true sharing of code, which is one of
the greatest promises of open source.  And it doesn't sound like either
party is willing.

Aaron


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/28/2000 4:06 PM, "marc fleury" <ma...@telkel.com> wrote:

> Indeed if the Avalon guy puts jBoss code in his tree and "contains" our work
> in his work then yeah.. that needs to be GPL.

Bingo. So, this is something that is a major problem for me.

> This is the mutations I was talking about... but the ASF decides not
> to mingle with 60% of the world's OSS codebase *deliberately*.

I'm sorry, but where exactly do you get that 60% number from?

The ASF decides not to mingle with X% of the worlds OSS codebase
*deliberately* for a reason. The reason is simply that the ASF does not
believe in putting excessive restrictions on source code to prevent what
people can and cannot do with it.

This is different than the GPL ideology which is based on a completely
whacked view of the universe that RMS decided to inflict on people.

In my mind, OSS is about simply getting credit for the work that you do, not
requiring people to either jump through hoops or give back their additions
or changes to my source code. I don't give a rats ass what people do with my
source code as long as they simply give me credit for my hard work.

Obviously you don't understand that concept yet. 7 years ago when I started
contributing to OSS projects, I didn't get it either.

In time, I have hope that you will eventually "get it". Good luck. I hope
that you don't get to badly burned in the process.

-jon


RE: [jBoss-Dev] Re: jboss on tomcat update

Posted by marc fleury <ma...@telkel.com>.
|> | but at the same time, you have a problem with the GPL being
|> |viral so you give exceptions for people to use JBoss. Instead, what you
|> |should do is probably be using the MPL license which will solve
|your needs
|> |without having to constantly grant exceptions to people.
|>
|> ???
|>
|> what 'exceptions'? we never granted 'exceptions'.  Please explain.
|
|The exceptions that you are granting is by allowing people who write EJB's
|for your server to allow them to not require them to be GPL'd as well. That
|is clearly an exception to the license. This is very similar to what Linus
|has done with Linux and binary kernel modules.

jon,

applications are *clearly* not covered by the GPL in J2EE (their work
derives from J2EE classes i.e SUN we are *NEVER* in the picture, not even
distribution).

Linux does

1- puts a "notice" (a.k.a FAQ) *outside* the "Terms and Conditions" to state
the obvious, that applications that use linux with "normal system calls" are
not derived from Linux (Oracle/apache derived linux?) and therefore not
covered... in our case we don't even have to make that obvious statement,
beans are derived from J2EE.

2- The ONLY exception he makes to the license is the copyright (the authors
and not the FSF)... we do the same (which means the code belongs to us not
the FSF so there is nothing the FSF can do it is not it's code, (you know
for those yelling "FSF will put you in jail" and all :))

At any rate the discussion on "modules" that make the server can cause more
confusion if you work from "impressions and vapors of the GPL" (tm), however
apps are clearly off-bounds.

|I write code for the ASF under an APL 1.1 license. The GPL and the APL 1.1
|are not compatible licenses and it is "illegal" for me to include GPL code
|within an ASF project. Period. Thus, I cannot take JBoss and
|include it with
|the Turbine Developer Kit because you have things under a GPL license.

Ok Brian explained that one to me the other day.  The reason is not so much
the licenses (as we see the two work together happily in the world today)
but the fact that you don't want to put GPL in your tree because of
propagation.

Indeed if the Avalon guy puts jBoss code in his tree and "contains" our work
in his work then yeah.. that needs to be GPL.

But understand it is a "decision" from the ASF not to include GPL code in
the tree for "purity reasons".  Many people integrate GPL all over the place
with other licenses without violation (i.e respect GPL all linux distros
do).  This is the mutations I was talking about... but the ASF decides not
to mingle with 60% of the world's OSS codebase *deliberately*.

interestin', an open license and a closed tree

marc


|Sigh, I feel like I'm repeating stuff to you again.
|
|-jon
|
|--
|http://scarab.tigris.org/    | http://noodle.tigris.org/
|http://java.apache.org/      | http://java.apache.org/turbine/
|http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
|http://www.collab.net/       | http://www.sourcexchange.com/
|
|
|
|


Re: [jBoss-Dev] Re: jboss on tomcat update

Posted by Jon Stevens <jo...@latchkey.com>.
on 10/27/2000 10:10 PM, "marc fleury" <ma...@telkel.com> wrote:

> | but at the same time, you have a problem with the GPL being
> |viral so you give exceptions for people to use JBoss. Instead, what you
> |should do is probably be using the MPL license which will solve your needs
> |without having to constantly grant exceptions to people.
> 
> ???
> 
> what 'exceptions'? we never granted 'exceptions'.  Please explain.

The exceptions that you are granting is by allowing people who write EJB's
for your server to allow them to not require them to be GPL'd as well. That
is clearly an exception to the license. This is very similar to what Linus
has done with Linux and binary kernel modules.

> |It is funny to me how you say that you are integrating our code which I
> |think is great, but the real issue is that we can't integrate YOUR code
> |because you choose to use the GPL license.
> 
> why not?  what exactly prevents you from integrating our work?  Please be
> explicit,
> 
> let's not work from hearsay and "impressions" of the GPL, the GPL is very
> explicit.

I write code for the ASF under an APL 1.1 license. The GPL and the APL 1.1
are not compatible licenses and it is "illegal" for me to include GPL code
within an ASF project. Period. Thus, I cannot take JBoss and include it with
the Turbine Developer Kit because you have things under a GPL license.

Sigh, I feel like I'm repeating stuff to you again.

-jon

-- 
http://scarab.tigris.org/    | http://noodle.tigris.org/
http://java.apache.org/      | http://java.apache.org/turbine/
http://www.working-dogs.com/ | http://jakarta.apache.org/velocity/
http://www.collab.net/       | http://www.sourcexchange.com/