[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: ALWAYS FREE?
From:       Simon Hausmann <shaus () uermel ! Med ! Uni-Magdeburg ! DE>
Date:       1999-09-14 16:31:40
[Download RAW message or body]


I don't really want to jump into the discussion, but:
Even the OMG itself copyrights CORBA interfaces...

example:
...
/*
 *  CosPropertyService
 *  Copyright (C) OMG
 */
...

Bye,
 Simon

> > > That's exactly my point: Since the idl file only represents
> > > an interface, there is no point in trying to impose a
> > > copyright on it. Except if you want to spread FUD about
> > > the conditions under which the interface can be used.
>  
> > Well, there is a copyright on it. That's automatic.
> 
> No. Interfaces are not copyrightable. If that would be the
> case, things like WINE would be impossible.
>  
> > The same thing happens with header files, which also define an interface,
> > doesn't it?
> 
> First of all, header files in C++ do not define interfaces.
> They contain all sorts of stuff like private variables,
> private methods, templates and inline functions. So if
> you include a C++ in your program, then the executable
> contains code from the header file. If this exceeds what
> is usually considered fair use, then you have to comply
> with the license conditions of the header. That's e. g.
> the reason why putting a template library under the LGPL
> is a bad idea (and in fact, libstdc++ is not under the LGPL).
> 
> Second, the purpose of a copyright notice in a header file
> is that it tells you about the conditions under which you
> can link with a library. For using a CORBA service, you
> don't link with a library (you may link with a library
> like libkded that wraps the service. In that case the
> respective library deserves a copyright notice). An IDL file
> specifies some kind of protocol between applications which 
> could even reside on different computers in different 
> countries. Copyrighting it would be like copyrighting http 
> or nntp.
> 
> Bernd.
> 

[prev in list] [next in list] [prev in thread] [next in thread]