Forum OpenACS Q&A: Response to GNU Public License and Mozilla Public License

Posted by Mat Kovach on
I think it has been proven that the only true way, unfortunaly, for GPL'd software to allow non-GPL software to run with it.  Seeing that, the FSF made the GPL rather compatiable with that.  This as I see it is a good thing.  Face it, OpenACS was developed so you could stop using ACS with Oracle (GPL'd code vs Close Source).

Given then, Close Source software will have problems with GPL'd code as it will be hard to _not_ make changes that break the close source software.  Just look at the Linux Kernel, yea there are binary modules but Linus does not promise the next version of the kernel will keep the compatability.  So you take a chance doing that.

There are plenty of "reason" people don't/won't release source code, legal, greed, etc.  But if the choice is closed source or nothing it can help a project continue and gain acceptance.

Of course, I don't think I'll ever like it, but I can accept it with cause and caution.

Now, the Mozilla Public License was developed for a specific purpose, so Netscape could release the code of a closed source product.  They were also going to produce/sell that product.  This obviously had many many legal issues and in that I see the Mozilla Public License as an excellent piece of work for people that need to do the same thing.  It is not good for a Open Source GPL'd project to go closed source (to any level).  I would have much rather seen aD goto a BSD type licence instead.

Netscape/Mozilla was basically saying they wanted to be Open Sourced but we can't because of the obligations associated with a closed source product.  aD has seem to say we want to move to closed source because we can't make money being GPL'd but we still want the benefits of being Open Source'd (this is my personal thoughts on the matter).

Obviously, mixing the two license is a recipe for disaster.

Being GPL'd allows somebody if they want to create an application (for example an ISO900X or CMM management system) that co-exists with OpenACS (or even ACS TCL) that requires NO MODIFICATION to that
base system, but run on it.  I'm not sure of the legal specifics of how it could be done though.  The MPL, as my bodyfeeder^H^H^H^H^H^H^H lawyer tells me, that would not be a clear cut possibility.

Just my thoughts.