"Frank Hecker" <[EMAIL PROTECTED]> wrote in message
[EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
> In this way you guarantee that you (and others) receive the same benefit
> from the new code that the author of that code received from your code.
> This creates an environment in which all code is shared among all, with
> "free riding" discouraged as much as possible.
Exactly!
> As I said, this is a perfectly valid view of the world to have, and it's
> basically the world view embodied in the GPL.
The difference is that I don't like the "viral" aspect of GPL, i.e. as per
MPL, you should be able to <use> my code or any derived code in your
application without being forced to publish your own propriatory code.
> However I think it is also possible to have a different view of the
> "derivative works" question. In this view, the author of the new source
> file (with the code for the subclass) has created something new that did
> not exist before, and could not necessarily have been predicted given
> the code which already existed. (After all, the author could have added
> new methods implementing entirely new and unique functionality,
> functionality not present in any form whatsoever in the original class.)
The s/he could create the class from scratch instead of becoming a
"free-rider". The point is that if s/he is saving time by using and
modifiyng my class, than these modifications should go back to the
community. That the license requirement for using this code in the first
place. If s/he doesn't like the conditions, then s/he is free to develop the
class from scratch and separately from the original code.
> In this view of the world, the code in the new source file wouldn't be
> considered simply a modification of your existing code, unless the
> author actually copied code text from your source files. Instead the new
> source file would be considered a new work, a separate act of creation.
> And as a new work, this new source file could potentially be licensed in
> a different way than your original code was licensed, if the author of
> the code chose to do so.
>
> IMO this is a different but equally valid way to view the "derivative
> work" question, and it is basically the view embodied in the MPL.
If this is the case, then in case of OOP code this makes MPL useless,
because it doesn't really give any advantage over BSD. It's maybe good for
COBOL programs, but it's totally inadequate for OOP code.
> In your case your views seem to be closer to those that motivated the
> creation of the GPL, which is why I suggested that the GPL might be a
> better license for what you want to do.
As I mentioned before, because of the "viral" effect, GPL is not an option
for me.
--
Michael Beck (Team JEDI) http://delphi-jedi.org
http://www.geocities.com/beckmi/delphi.htm
