Hi,

"Rickard �berg" wrote:
> Ole Husgaard wrote:
> > 1) jBoss cannot include the Tomcat code and distribute
> >    the combination without breaking the APL license of
> >    Tomcat.
> 
> Which voids all bundling efforts like OpenJoda, or Aarons. Not good. EJB
> on its own is quite weak.

An EJB server implementation on its own is of very little
use. But it can be GPL, and I think it can be made to
stand on its own.
You may ask: Why would we want that if it is of very
little use?
The reason is: If is is GPL and can stand on its own,
we are free to aggregate, as GPL places no restriction
on "mere aggregation".

But like Peter Donald pointed out, is must be free of
license conflicts and able to stand on its own. For
example we should not hardcode the call to start the
embedded Tomcat into jBoss; it should be in a
configuration file (in a sensible manner so that users
can configure to start any web application server that
adheres to how jBoss starts embedded web application
servers) and reflection should be used for the startup
call.

> > We should concentrate on the facts:
> > 1) Copyright legislation.
> > 2) APL license.
> > 3) GPL license.
> > 4) Are any of the three above violated, and if
> >    yes: How, and what are the implications?
> > These three things are the _only_ ones that a
> > lawyer would look at if he was preparing a case.
> 
> I am personally more concerned with what various uses of jBoss are
> possible. Currently it is not possible to make a J2EE server with jBoss
> and Tomcat, and that worries me.

Please do not worry. I am pretty sure we can get this
sorted out in a way that everybody is happy about.

> How many want GPL (for whatever reason) and how many would be ok with
> APL (or similar, again for whatever reason)?

I think that licences like BSD license and APL allow
too much. But I am not religious. The Library GPL
could be closer to what I think most jBoss people
want.

But we should really try to get the problems sorted
out first. I worry a lot less now than I did two
days ago.


Best Regards,

Ole Husgaard.

Reply via email to