Paul Winkler <[EMAIL PROTECTED]> wrote:
> Michael Stutz wrote:
> I've looked at the Open Publication license and so far I like this
> new one of yours best, though I'm not sure about the name ("Design
> Science License?" Why that, pray tell?)
Heh. I'm totally open to a better name! At first I thought of using
something like "Copyleft License" or "General Copyleft Public License"
or something along those lines, but those names seem _too_
generic. Then I thought about the GNU GPL's name, which was
idiosyncratic to the FSF organization (I wonder now if they ever
thought people would use it for non-FSF projects?), but it caught on
and people used it. But yeah, if there's a more appropriate name for
this, it should change!
As for why the name: "Design Science" is "the art and science of doing
more with less to ensure the best possible outcomes for humans in
Universe"; it's what had originally attracted me to the free software
movement and to copyleft -- the idea that people can cooperate and
communicate honestly and effectively regardless of obsolete sovereign
laws like copyright. So it seemed to me that a copyleft license that
anyone could use for anything would be an ultimate step toward design
science revolution: http://www.bfi.org/design_science.htm
> Cool. This addresses the problem with applying the GPL to music.
> After all, if you make an audio CD of MIDI music, it's impractical
> to put the MIDI files on the same CD. (Although I don't think it's
> impossible.)
It's not impossible. But you're right that it shouldn't be
mandatory. And I wonder how much longer audio CDs will be in use.
> The same cannot be said for music.
It seems that what the "source" is for any work always depends on a
few things: what language or encoding the work is in and how the
author composed it.
It's possible to write a software program in machine language. It's
not a very practical way especially as program complexity increases,
but it's possible. But if an author writes a program as a binary, the
source is that binary.
I think you're right about MP3, that it isn't a source form (in most
cases; again it depends on how it was composed). The source ought to
be the AIFF or WAV file from which the MP3 was made. When I make an
mp3 of an analog tape recording, I first record the WAV or whatever,
and then burn an mp3 from that. So the WAV recording would be the
source. But if I obtain an mp3 file from somewhere and edit it
somehow, that mp3 would be the source; or if I write a program for
composing music that somehow edits and works on mp3 files, it's
possible they could be the source. (The idea of those examples is to
clarify all of this. So they should probably change to make a better
clarification of all this.)
> I've given some thought to this and I don't see an obvious solution
> for every case, since in many cases the real "original composition"
> may take orders of magnitude more disk storage and network
> bandwidth.
Compare it to source code. What if a programmer has a huge collection
of macros, and an elaborate programming environment with years of
settings and preferences and things to make programming easier for
that programmer. Still nothing but the ASCII text compileable files
constitute the source -- the final product, the thing that the
programmer was trying to "say"; you don't have to somehow include all
of his drafts, notes, sketches, pages of illustrations and flowcharts,
revisions of the source, encode his knowledge on the subject, his
programming macros, etc.
So what about this test for what the source is: as long as anyone else
(with reasonable equipement such as compilers, libraries, etc) can
take it and turn it into what you have "said" or communicated in
object form, as well as take it and make it "say" soemthing else, and
then put that in object form.
> But if I record a project in ProTools, are the ProTools files my
> source code? Anyone who wanted to use them would have to either own
> ProTools or reverse-engineer its soundfile format and edit
> lists.
This might be like a software program that's written in a computer
language which only has one proprietary compiler, written by the
company who invented the language.
> These questions are getting more and more absurd, but the point I am
> trying to make is that the answers are not obvious, and I am
> concerned that this could be a problem if you ever needed to take
> anyone to court over a violation of this license.
And it just might happen to someone. Has it ever happened with
software? It seems like most software licenses are enforced by
"embarassment," making the violations public. I don't think anyone's
gone to court (yet) over an open source violation...
> Must I put 650 MB on a web or ftp site in order to comply with the
> license?
Some things are going to have much larger "source" material than
others, like images and music. It won't be practical to include that
with the object form. What about making it available by postal mail,
like as with GNU software? You can charge transportation and media
costs. Current linux kernel sources are 65MB+, for a 429k kernel
image.
> One possible remedy: What if there was a condition (d) like so:
>
> (d) A written offer for obtaining the Source stored on a physical
> medium, at a nominal fee adequate to cover the cost of creating and
> shipping the physical medium. This option is valid only if the
> Source is of sufficient size as to place an undue burden on the
> author(s) or distributor were the Source to be placed in a
> publically-available location such as a website.
I like this. Something like this probably should be included.
