Don Armstrong <[EMAIL PROTECTED]> writes: > On Sun, 19 Mar 2006, Måns Rullgård wrote: >> Don Armstrong <[EMAIL PROTECTED]> writes: >> > Not just linking; it's the creation of a derivative work of a >> > GPLed work. Frankly, I don't see how you can argue that cdrecord >> > is not a derivative work of the GPLed part of cdrecord and the >> > build system. >> >> I disagree. The final executable is no more a derivative of the >> build system than it is of the compiler. After all, no parts of the >> makefiles end up inside the executable. > > The makefiles direct the assembly of the executable, just like the > source code directs the operation of the compiler. [And indeed, some > question as to whether some part of the executable is a dirived work > of the compiler exists as well; luckily there are exceptions in the > licences of gcc to deal with this case.] > > There are multiple different ways that the compiler and assembler can > be directed by the makefile; quite a large number of them will produce > an operational executable.
Given only the source files, writing a makefile that will produce a working executable is fairly simple. I see makefiles as more of a convenience than a necessity to build a program. You might as well argue that every program in Debian is a derivative of apt and the package descriptions, since the former uses rules in the latter to decide what to install. Again, I say this is just a convenience that saves users the time to find out and install the dependencies manually. -- Måns Rullgård [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
