Gennaro Prota wrote:
On Thu, 27 Jul 2006 19:46:10 +0200, Markus Schöpflin
wrote: Huh, boost is not free software? AFAICT, the license allows all the four freedoms. See http://www.gnu.org/philosophy/free-sw.html for a list of those. So why should boost not fall into the category of free software?
Hi Markus,
not sure whether you replied after or before my auto-followup :-) In any case, yes, it allows all four freedoms and thus qualify, by definition. However it also allows the freedom to "remove freedom" on a specific derivative work: it is non-copylefted free software.
Before, but with a non-subscribed mail address, so it took some iterations on my side and moderator approval to get the reply to the ML. :-) Anyway, you are of course right by saying that boost qualifies as non-copylefted free software. If this is more or less than just free software is a philosophical question. I was just mystified by your original comment claiming that boost is *not* free.
This can be seen as "it gives more freedom", as Brian suggests, or "it doesn't preserve freedom", according to one's one beliefs and/or logic. As a matter of fact, that allows cases such as X11, which has non-free versions being the only ones that work on some hardware (making it non-free for users of that hardware).
Yes, but it also allows for cases where (like in our case) we are using boost in commercial closed source products without too much legal hassle. This would have never been possible if boost was copylefted. The benefit for the community in this case is that boost is regression tested on a more exotic platform (Tru64/cxx/gcc) on a regular base. This may be a bit exaggerated, but when boost would be copylefted, there possibly would be no up to date support for Tru64. Markus PS: Someone recently posted a link to an excellent article regarding these issues to the boost developer mailing list. See http://www.ipinfoblog.com/archives/Open%20Source%20Legal%20Issues.pdf