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

Re: ITP:Bug#460591 - Falcon P.L. license

On Wed, 19 Mar 2008 16:00:31 -0700 Sean Kellogg wrote:

> On Wednesday 19 March 2008 03:10:07 pm Francesco Poli wrote:
> > I don't think that copyright laws give you the right to control
> > distribution of "Scripts", that is to say, works written in your
> > programming language.
> > For instance, I don't think I need permission from anyone in order to
> > write (original) C++ code and distribute it as I like.
> This is an interesting claim and I think is far from clear cut. If I develop a 
> comprehensive grammar and syntax for a language (programming or otherwise) 
> what elements are required for others to use "speak" it and to what extent 
> can I protect/restrict the use of those elements? The first obvious answer is 
> that the language is an idea, and thus patentable. I don't think that's a 
> point of contention here, but worth noting that one can certainly restrict 
> the use of a language via a patent.

This is only true in those (insane) jurisdictions where "ideas" are
patentable.  That is to say, where software patents and the like are
valid.  Examples of those jurisdictions are, IIUC: the USA (but there
are plans to push for a political change in the USA), Japan,
Australia, ...

Anyway, as you yourself acknowledge, I was talking about copyright laws

> But, if we keep our analysis to just copyright protection, I'm still not sure 
> it's a slam dunk depending on factors. Here I'm thinking of libraries which 
> may be necessary for the script to run... like libc. Sure, I can just write C 
> code, but it's pretty useless without libc...

For the sake of clarity, I was talking about writing and distributing
(original) C++ source code.  I didn't (on purpose) consider the
distribution of compiled *and linked* code.

It is my understanding that writing source code intended to link with a
library, does not, per se, create a derivative work of a specific
implementation of that library, as long as you only adhere to an API.

In your example, please note that there are many differently-licensed
implementations of the C standard library.  If I write C code by only
adhering to the standard C library API, then I don't think I'm
restricted by the license of any libc implementation... 

> and if I incorporate a version 
> of libc with a restrictive license, then I'm certainly going to have to 
> comply with the copyright restrictions of that license.

Let's not open the can of worms of static/dynamic linking versus
derivative work creation, please!  ;-)

> There's even the question of how someone goes about learning the language... 
> presumably by example (that's how I've learned most other languages).

By reading a book which explains the grammar, syntax, and semantics of
the language?  The book will probably include examples, but I don't
think that copying one line of code from an example (especially when
that line is the only good way to use an instruction or to call a
library function) creates a derivative work...
Hence, in most cases, your code won't be a derivative work of the
examples you once studied.

> Can I 
> create a work that is not derivative of those examples and thus under the 
> preview of copyright law?

Is this e-mail message a derivative work of the example texts I once
studied on my English grammar and conversation textbooks?
I don't think so...
Well, at least, I hope it's not!  ;-)

> This would have made a fun topic to write about in law school :)

IANAL, nor did I attend law school, hence I can only try to imagine how
fun it would have been...

My other disclaimers still apply: TINLA, IANADD, TINASOTODP.

 New! Version 0.6 available! What? See for yourself!
..................................................... Francesco Poli .
 GnuPG key fpr == C979 F34B 27CE 5CD8 DC12  31B5 78F4 279B DD6D FCF4

Attachment: pgpJTSkGYBdGR.pgp
Description: PGP signature

Reply to: