John Halton wrote:
1. One could argue that objections to the ASP "loophole" come down
to a reluctance to accept the implications of the "no
restrictions on use" aspects of free software: "How dare people
make money out of the software I've written?"
I don't know about the AGPL. I stopped trying to grok the entirety of any FSF license, but I am still forced to read them now and then because whether I like it or not, the GPL gave strength to the FOSS movement and it still plays a very large role.
Anyway, I feel you miss the point where someone who licenses their software under a license with an ASP-fix clause does not want to prevent their consumers (the service providers) from making money, any more than Linus Torvalds who uses GPL for his software wants to prevent Red Hat and SuSE from making money. They only want the *software* to be open and free. I am surprised you could have such a misunderstanding as stated by you above.
But even if we take position #2, that still leaves the fundamental problem as this: the AGPL makes it possible to add new andincompatible licensing restrictions to GPLed software.
And I thought I was so careful in labeling this thread non-specific to AGPL. AGPL was only an example. I mentioned it because it is poised to becoming very widely used by virtue of being blessed by the FSF.
My question that started this thread was whether a simple ASP-fix clause would make a work non-DFSG-free. The licensing terms *I* am presenting for discussion is a Sleepycat+ASP-fix license which I have already outlined on this list.
a de facto threat to the GPL software ecosystem, as we could see more and more useful software moving from the GPL world into the AGPL world.
Are you saying that many companies are now making money by not freeing and opening the FOSS, private modifications and derivative works that they ostensibly "privately use" and "not distribute" for their network services and those companies could face financial troubles if they are forced to free and open up their private modifications and private derivative works ("private" meaning "non-distributed" in the pre-AGPL sense of the term "distribute")?
If so, I agree, and seeing as the GPL v3 allows works under it to be combined with works under the AGPL v3 and be combinedly distributed under the *AGPL* only, the FSF is apparently in support of this world-view. FOSS commerce will be redefined, I guess. But really, I don't see the Linux kernel or many other GPL-ed apps going under the AGPL, seeing as the pure-desktop usage market is still wide, and is not going to disappear in the foreseeable future.
Shriramana Sharma.