| |||||||||||||
Sadly ...Sadly ...Posted Mar 14, 2012 5:45 UTC (Wed) by ldo (subscriber, #40946)In reply to: Sadly ... by geofft Parent article: Gnuplot 4.6 released
The thing you link to satisfies the FSF Free Software Definition... No it doesn’t. May I draw your attention to Freedom 3: “The freedom to distribute copies of your modified versions to others”.
(Log in to post comments)
Sadly ... Posted Mar 14, 2012 5:56 UTC (Wed) by geofft (subscriber, #59789) [Link] Yes, I see it. The license explicitly grants me "permission to distribute binaries produced by compiling modified sources", which satisfies the most pedantic reading of that rule, in that it applies to the software and not its source code. A more useful, non-pedantic reading would acknowledge there is no practical distinction between distributing modified code and distributing patchfiles. The cases where there is a distinction (and there certainly are such cases!) are those like Pine, where source patchfiles could be distributed, but modified binaries could not (and even unmodified binaries could not be generally redistributed). Pine was nonfree, not because of the restriction that you had to distribute modified source in the form of original source plus patches, but because of the constraints regarding distributing the binaries -- what Freedom 3 refers to as "your modified versions".
Sadly ... Posted Mar 14, 2012 11:03 UTC (Wed) by nye (guest, #51576) [Link]
There are other practical distinctions. For example, this license does not allow you to re-use any code in another project, unless you structure that project in terms of a modification to gnuplot.
It's explicitly designed to minimize code re-use, which to my mind makes it non-free in effect. That it technically satisfies the definition seems more like a bug in the definition to me.
(Also, it requires the modifier to attach their meatspace address to any modifications, which is a fairly nasty restriction for many, though not in itself non-free.)
Sadly ... Posted Mar 14, 2012 18:04 UTC (Wed) by geofft (subscriber, #59789) [Link]
Good point, I had failed to consider reusability in other software
In fairness I think the FSF definition does a bad job of mandating it / making it clear that this is important (as do the Debian, OSI, etc. ones), but it's certainly something I personally think is very important to "free" software.
|
|||||||||||||