Quote:
Originally Posted by gberardi  I will agree that when it comes to someone wanting to use the source code in the way they want, the GPL is more restrictive than the BSD license.
However, the GPL's restrictions are there to enforce certain freedoms and ensure that they remain free. The MIT/BSD/ZLib licenses can't guarantee those freedoms since the author of derivative code can do whatever he/she wants, as you said. This difference is why people choose the GPL over those other licenses. They want to make sure that future users/developers will have the same freedoms that the GPL guarantees. |
Here's how I see it. If I write some code that I want to release to the public to provide the biggest benefit to anyone, regardless of how they want to use it, then GPL is not a good choice. Why? Because I wrote my code and decided to distribute it for free, and if people want to use it in their software without releasing
their code then why shouldn't they be able to? Yes they can technically improve my code and not release these improvements, but this is their choice.
They put extra work in to improve the code, so they have the right not to release their investment.
The best example I can give of where MIT/BSD/ZLIB licensed code made a great contribution is Mac OS X. If FreeBSD was licensed under GPL, the current Mac OS X wouldn't exist. Even Microsoft used some FreeBSD code in Windows. And while Apple obviously didn't release OS X for free, they did release Darwin (the core of OS X) under an FSF approved open source license. Not forcing people to release their code doesn't prevent them from doing it anyway, and under less restrictive terms too. The bottom line is that I don't believe that any sort of control over other people via a license is true freedom, regardless of the intentions. Eric S. Raymond wrote a nice article which sums up my feelings on the subject:
Eric Raymond: Freedom, Power, or Confusion?