Which Python web framework is most like Ruby on Rails?

Paul Rubin http
Wed Dec 21 05:31:27 EST 2005


"Ben Sizer" <kylotan at gmail.com> writes:
> Unfortunately, that doesn't really satisfy the GPL's concerns. The work
> arguably "contains or is derived from" Karrigell,

I don't see that.  The web app gets run by Karrigell like a CGI script
is run by Apache, like a Linux app is run by the Linux kernel.  The
Linux kernel is GPL but that doesn't mean every Linux app that calls
through the kernel through its documented API is affected.  The Linux
kernel GPL only comes into play if you start modifying the kernel, not
writing programs that use it.  

> which is explicitly covered in section 2b of the GPL. If you start
> excluding key clauses from the GPL, then there's little point using it.

There should be no exclusion.

> To cut a long story short, and to avoid quibbling over the details of
> how a license designed with the low-level mechanics of C-style programs
> and libraries in mind actually applies to a language like Python with
> very loose coupling, I'll just say that this sort of situation is
> exactly what the LGPL exists for. 

The LGPL has fallen into some disfavor at the FSF, and I don't see
Karrigell as a library.  The application is not an integrated blob
combining Karrigell and user code, from what I can tell.  Rather, the
user code is run under Karrigell's control, like Karrigell itself
is run under the Python interpreter.

> I would suggest the author adopts the LGPL as a good compromise
> between the community benefits of GPL and the user benefits of, say,
> BSD or zlib licenses.

But that doesn't gain the community benefits of the GPL since it
foregoes the possibility of users improving Karrigell by incorporating
other GPL'd code into it and distributing the improved version.



More information about the Python-list mailing list