I find for many things the BSD License, leaves a bad taste in my mouth. Not all things of course but some. To me it allows theft of work product. I know this is not the legal interpretation, it is a philosophical one and CMD is just as guilty of this theft as others in the PostgreSQL community including EDB and GreenPlum. The theory behind the BSD license is a good one. The great people contribute because it is the right thing to do. The good people contribute because they are compelled by the greater force of the community. The bad people? Well they steal it and call it their own. The best scenario is that a contributor is compelled to contribute per the value of the content. That content could be code (PostgreSQL). Reality dictates that there is a general level of community coercion that takes place and that the great people are few, the good people are some and the bad people are many. The LGPL seems to present the most reasonable flexibility of Freedom (which BSD provides) and community contribution (which the GPL forces). It is reasonable that as an author of software I can close that software. It is reasonable that a user may chose not to use that software because I chose to close it. It is not reasonable to force someone to open their software. It is absolutely not reasonable to close source modifications made to software that you don't own without giving back to the community (or author). The LGPL provides us with the protection to enforce that code stability. Should I chose to author and compile a piece of code on Linux, I can close source it because glibc is LGPL. However if I modify glibc itself, I have to give the change back. To me is that is a definition of fair. Lastly, no I don't think PostgreSQL should change their license (I don't think they could anyway). The BSD has worked well for PostgreSQL. I didn't say BSD was bad in all cases.