[BioPython] licensing

Jan T. Kim kim@inb.mu-luebeck.de
Wed, 22 May 2002 13:04:06 +0200


Hi all,

On Tue, May 21, 2002 at 12:31:25PM -0600, Andrew Dalke wrote:
> Christopher Lee:
> >I realize this may be off topic, but I'm worried that you may be spreading
> a
> >misconception about the (L)GPL.  Copyright law specifically *doesn't* apply
> to
> >ideas.  The LGPL would prevent you from copy/pasting (or or translating)
> code
> >into your proprietary code, but you would be free to take ideas in LGPL
> code and
> >use it in proprietary work.
> 
> First, IANAL so you may be right.

> There's always the possibility of accidental copying of code, where
> since I know how one chunk of code works, I may 'rewrite' it but where
> the new code is essentially the same. [...]

Not being a lawyer either, my understanding here is that Christopher
is right in principle, so the lawyers have chosen to create confusion
about what an idea (or concept etc.) is, and what code is, so they can
effectively "copyright" ideas -- which is paradoxical, since ideas
cannot be copied at all.

Further addressing this issue is too much off the topic of this list, but
one should notice that it's really not the LGPL which is to be blamed
for any trouble along the lines outlined by Andrew, but it's the mess
created by (international) lawmakers and lawyers.

> And finally, this issue of taint/clean doesn't address my concern that
> people who want to develop software which is incompatible with the [L]GPL
>  -- and that includes some free/open source licenses -- are unable to
> do so and instead waste time rewriting the code from scratch, without
> even the ability to scavange for parts.

Not necessarily. The LGPL says

    6. As an exception to the Sections above, you may also combine or link
    a "work that uses the Library" with the Library to produce a work
    containing portions of the Library, and distribute that work under
    terms of your choice, provided that the terms permit modification
    of the work for the customer's own use and reverse engineering for
    debugging such modifications.

All this effectively does is preventing you from selling LGPLed code as
though it was yours. I cannot see how anyone could want to be incompatible
with this concept without being incompatible with the concept of fairness
at the same time. With the LGPL, the "taint of freedom" cannot spread
out into any "clean" code. (Personally, I think that the metaphorical
attibutions of "tainted" and "clean" are wrong here, I chose them this
way for consistency with the preceding discussion only.)

Greetinx, Jan
-- 
 +- Jan T. Kim -------------------------------------------------------+
 |             email: kim@inb.mu-luebeck.de                           |
 |             WWW:   http://www.inb.mu-luebeck.de/staff/kim.html     |
 *-----=<  hierarchical systems are for files, not for humans  >=-----*