> this nice gpl,lgpl and commercial discussions...
> i repeat a statement already mentioned in this thread:
> don't trust anyone telling you something about how to behave concerning law
> just trust your paid lawyer.
Well, listening to a lawyer is good, but most lawyers don't pay the bill
and suffer the consequences if the court decided against you in a dispute.
So in the end it is down to the organization or individual to decide
what to trust and what not to trust.
> AS LONG AS you link dynamically to all lgpl code.
Dynamic or static linking is no issue here, LGPL allows both.
Section 6 applies to what has been discussed here ie LGPL library and closed
source application delivered as an executable.
There it says (6a) that if you distribute statically linked application
you also need to distribute it in un-linked form so that the user
can (re)link it with a modified (improved or different) version
of the library, if they want. And 6c says that you don't even
have to distribute the un-linked form, it is enough if you
promise to deliver it on request.
6b refers to dynamic linking allowing distribution of the library alongside
with a dynamically linked application code, but it is worth noting that
section 5 clearly spells out that an application compiled against the
library falls outside the scope of the license and thus there is nothing in
the LGPL license to stop distribution of dynamically linked application code
in any shape or form as long as it does not contain the library.
Worth noting is that the last two paragraphs of the section 6 require the
distribution of the tools and libraries required to (re)link the application
unless they are normally distributed with the OS.
This may cause problems if you use libraries or tools that do not allow
redistribution. For example I don't think Windows comes with a linker
so static linking might require you to distribute the linker which
might not be possible if you use M$ tools. On the other hand if you
use Free tools such as MinGW you might be able to evoke 6c on them.
Qt-interest mailing list