[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: On interpreting licences (was: KDE not in Debian?)



William T Wilson wrote:
> 
> On Mon, 7 Feb 2000, Raul Miller wrote:
> 
> > > b/c "executable work" as written in the quoted sentence above refers
> > > to the executable work as it is being distributed, not as it exists
> > > at run-time).
> >
> > You're claiming here that even though Qt must be linked with
> > kghostscript that the executing program doesn't contain Qt?
> 
> The executing program isn't relevant because the GPL doesn't specify the
> conditions the program can be run under.  It only specifies the
> conditions the program can be distributed under.

The program as _distributed_ contains portions of Qt, even if dynamically
linked.

Macros. Data structures. Method definitions. All are parts of Qt
necessary for compilation, and which are included in the final
executable.

> > > I.e., the GPL does distinguish b/w dynamic and static linking.
> >
> > It doesn't even use the term "linking" in the terms of the license.
> 
> I think it does, by implication if not directly.  If you link statically
> with a proprietary library which is not part of the operating system
> then you cannot distribute under the GPL.  But you can if you link
> dynamically, because you aren't distributing any proprietary code at
> all.  You're just assuming that the required proprietary code will
> already be on the target system.

Dynamic linking has been the exception rather than the rule for most of
the history of computing. Most libc's, even proprietary, allow
distribution of static binaries.

-- 
| Jeff Teunissen -=- Pres., Dusk To Dawn Computing -- d2deek at pmail.net
| Disclaimer: I am my employer, so anything I say goes for me too.     :)
| dusknet.dhis.net is a black hole for email.    Use my Reply-To address.
| Specializing in Debian GNU/Linux         http://dusknet.dhis.net/~deek/


Reply to: