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

Re: wanting to package wpoison



In message <20020909201637.GF27645@morgul.net>, you wrote:

>On Mon, Sep 09, 2002 at 10:00:36PM +0200, martin f krafft wrote:
>>=20
>>   Redistribution and use in source form ONLY, with or without modificatio=
>n,
>>   is permitted provided that the following conditions are met:
>
>"source form ONLY" is not compliant with "...must allow distribution in
>source code as well as compiled form."  See section 2 of DFSG.

Ummmm ... the item in question, Wpoison, is a Perl script.

Generally speaking, for Perl, source == compiled.  At least that's
one way of looking at it.

I mean yea, there _is_ a ``compile to binary'' option for the Perl
Parser/interpreter, but I mean really, how often is it used?  How
many Perl thingies are ever (or have ever been) distributed in
true ``compiled'' form?  Like basically none?


Anyway, for whatever it's worth, I think your ``source as well as
compiled'' condition is your DFSG is downright silly.

I mean what's the practical difference between (say) distributing a
compiled binary of program `X' as opposed to (say) distributing a
shar file called X which, when invoked, stashes the source code for 
(say) X.c in a temp file, invokes the C compiler on it, and then
copies the resulting binary back over the original executable shar
file `X'?

What are you going to do?  Prevent yourselves from distributing a
program that's been packaged in that way, just because it doesn't
conform precisely to some abstract notion you have of what does
and doesn't constitute ``compiled''?

Why not just tie a boat anchor around your own necks?  That would make
about as much sense.


If you need a ``compiled'' binary, then create one, from the source
on the distribution media, on the fly, as the user is first installing
(or upgrading) the whole Debian system.  Make compilation part of the
system install process.  You can do that, can't you?

But just make sure that the distribution media contains the real and
honest-to-goodness source code... the stuff that a human being could
maintain and/or adapt, if necessary.

I think that this way of doing things is maximally consistant with
RMS's well known views regarding the free accessability of source
code.  Wouldn't you agree?

It also happens to be consistant with _my_ wishes (with respect to
Wpoison) as well.


Regards,
rfg


P.S.  Speaking of which, now that I think of it, I have just realized that
_my_ copyright conditions (for Wpoison) are probably not sufficiently
clear, and that I will be needing to revise them.

In theory, somebody could ``distribute source'' (thus satisfying my copy-
right terms) and yet do so on some media that require some ``secret'' and
legally-protected encryption scheme/mechanism in order to get the stuff
off the media in a way that would be at all useful.

And if they did that, of course, the DMCA would prevent any attempts (by
U.S. citizens/residents anyway) to ``break'' the encryption in order to
really get at the actual source code... i.e. the source BEFORE it has
been compiled (during the install).

I wouldn't like that.  Not one bit.

So I guess that soon I will amend my copyright terms to say that Wpoison
may only be distributed in source form AND that it MUST NOT be distributed
via any media which makes use of any form of encryption or copy-protection
scheme (or schemes) which either is covered, or which may be covered under
the DMCA.

(That will be my small way of saying to the world that the DMCA sucks.
Maybe you folks would like to join in with your license terms too.  And
if three... THREE people do it, then they'll think it is a movement... :-)



Reply to: