The Python Papers Edition One

[Paul Boddie]

IMHO your licensing terms are fine; you don't need to switch from the CC
license. Just avoid the term "free as in freedom", since the Free
Software Foundation has assigned that phrase a very specific meaning.

Agreed. It should also be noted that Debian - amongst the strictest
with regard to software and content licensing - do not regard the CC
licences as being "free as in freedom":

http://en.wikipedia.org/wiki/Creative_Commons_licenses
http://people.debian.org/~evan/ccsummary.html
http://people.debian.org/~evan/draftresponse.txt

Consequently, it may be appropriate to remind authors to also make
their works more widely available under a licence that may permit
further distribution, if that is desirable.

Paul

P.S. I still don't really understand why the FSF unreservedly
recommends the Free Art licence when it has a "choice of law" clause
that has prevented things like the GPL-compatibility of other licences
in the past.

 

[Carl Banks]

Just avoid the term "free as in freedom", since the Free
Software Foundation has assigned that phrase a very specific meaning.

Bah. FSF is not an arbiter of the language. People whose idea of
"free" differs from FSF's still need to differentiate it from the
monetary sense of the word free.

Carl Banks

 

[Jerry Hill]

My problem with this is that I can't use anything in your publication
when working on commercial software. If I were to derive code from
something in the Python Papers, my understanding is that I would be
obligated to release it under a Creative Commons license. In fact,
even if all I do is read an article and then incorporate concepts from
it in my code, my understanding is that I may be creating a derivative
work.

Since the code that I write for work belongs to my employer, and may
someday be sold, I need to be careful about the licensing issues. They
might not be very happy with me if I wrote software for them that
ended up being encumbered with a license they didn't like. I don't
think there's much of a problem with Issue 1, since I don't think I'd
end up copying the code in the 'Python Coding Idioms' article, and
MontyLingua is GPL software and thus has its own licensing issues for
commercial software. Still, if there was a useful article on, say,
database or web interfaces, I would have to avoid it. If I have to
avoid using any articles that might actually be useful in my
professional life, I feel like I should probably just avoid the
journal all together. If I've misrepresented the relevant copyright
issues, I'd be happy for someone to correct me.

 

[Steven D'Aprano]

And how do you think this is different from any other publication? That
Python Papers is under a CC licence is a red-herring.


Again, how is this different from any other publication? Unless you only
read public domain publications, anything you read is copyrighted, and
your arguments apply just as much -- perhaps more so -- depending on the
licence of that publication.

 

[Maurice LING]

As Steven mentioned -- anything you can read is copyrighted. The
difference is whether is the copyright effective or enforceable. What do
I mean by this? Without copyright, there will not be plagarism. Ask
yourself this question, can you copy William Shakespeare's MacBeth and
submit it as a literary work for a Master of Literary Arts degree? I
believe the candidate will be expelled from university. William
Shakespeare's MacBeth is still copyrighted work but not "enforceable"
because it is pre-1900's work and the author had been dead for more than
50 years. Similarly, works in public domain are still copyrighted --
academically, using work in public domain without attribution (giving
credits in the form of citations) is still plagarism.

This means that everything you had read since the days of "ABC..." are
copyrighted. That includes all codes you've seen in colleges etc etc. I
am afraid that to avoid copyright altogether, as far as your work is
concerned, you might have to seclude yourself in some pacific islands
and re-discover mathematics and computer science all over again from 1 +
1 = 2, and 2 + 1 = 3, and so on. Even so, patents will still get you at
the end.

In copyright, there is fair use. There is no way of avoiding it totally
-- how many ways are there to write a list comprehension?

Copyright just says attribute credits when you use someone else's work
within the limits of fair use; otherwise you might have to pay for it in
the form of a licence, subject to the copyright owner. I believe you've
done all these in college when writing your essays.

I believe in most cases, a simple declaration like "This function is a
re-implementation (or adaptation) of that found in <some periodical's
title, year, and page number>" will suffice.

Have you not read "The Python Cookbook", in book form or from the
website? How do you attribute credits when you are using the codes?

Cheers
maurice

 

[Robert Kern]

As Steven mentioned -- anything you can read is copyrighted. The
difference is whether is the copyright effective or enforceable. What do
I mean by this? Without copyright, there will not be plagarism. Ask
yourself this question, can you copy William Shakespeare's MacBeth and
submit it as a literary work for a Master of Literary Arts degree? I
believe the candidate will be expelled from university. William
Shakespeare's MacBeth is still copyrighted work but not "enforceable"
because it is pre-1900's work and the author had been dead for more than
50 years. Similarly, works in public domain are still copyrighted --
academically, using work in public domain without attribution (giving
credits in the form of citations) is still plagarism.

This means that everything you had read since the days of "ABC..." are
copyrighted. That includes all codes you've seen in colleges etc etc. I
am afraid that to avoid copyright altogether, as far as your work is
concerned, you might have to seclude yourself in some pacific islands
and re-discover mathematics and computer science all over again from 1 +
1 = 2, and 2 + 1 = 3, and so on. Even so, patents will still get you at
the end.

In copyright, there is fair use. There is no way of avoiding it totally
-- how many ways are there to write a list comprehension?

Copyright just says attribute credits when you use someone else's work
within the limits of fair use; otherwise you might have to pay for it in
the form of a licence, subject to the copyright owner. I believe you've
done all these in college when writing your essays.

I believe in most cases, a simple declaration like "This function is a
re-implementation (or adaptation) of that found in <some periodical's
title, year, and page number>" will suffice.

Have you not read "The Python Cookbook", in book form or from the
website? How do you attribute credits when you are using the codes?

I'm not going to go through this point by point, but nearly everything you've
said is wrong.

--
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless enigma
that is made terrible by our own mad attempt to interpret it as though it had
an underlying truth."
-- Umberto Eco

 

[Fredrik Lundh]

If anyone has any good ideas for how to cope as a publisher with these
difficulties, I'm all ears.

has any of the format zealots posting to this thread actually
volunteered to produce any material for your publication? if not, I
suggest ignoring them. any bozo with a keyboard can contribute stop
energy to a project, but at the end, it's always the people *doing*
things that matters.

</F>

 

[Paul Rubin]

Your email indicates a possible concern that we are doing something
untoward -- this was not at all intended, nor is it true.

Although you might not have intended it, I feel it is still true. You
are misappropriating terminology ("free as in freedom") from the free
software movement that means a work is licensed in a way that meets
some specific criteria, that the license you're using does not meet.

 

[Paul Rubin]

I thought I just had. In what way does the statement "Yes, it's true
that you can't resell copies of The Python Papers for personal profits,
but you may derive from it, reproduce and propagate it" not provide
such a revision and clarification? Seriously, let me know what exact
statement you feel needs to be made, and I will endorse it accordingly
if it is accurate.

You should not use the phrase "free as in freedom". "Free as in beer"
is more accurate for the NC licenses.

For my part, I don't see that there are ethically serious restrictions
on the freedom of use of the information contained within The Python
Papers.

There is a big rant against the CC-NC licenses on Kuro5hin:

http://www.kuro5hin.org/story/2005/9/11/16331/0655

Call it "mostly free" if you like.

That is reasonable and if you're going to stick with the NC license,
it would be cool if you also were to descibe TPP as "mostly free", or
maybe "gratis" (as opposed to "libre") in your descriptions. "Gratis"
is sometimes translated "free as in beer", in contrast with "free
as in speech" or "free as in freedom".

There's no such thing as complete
freedom of information anyway, and we have done the best we can.

Well, others including the Wikimedia projects (which use GFDL) and the
PLoS journals (www.plos.org) (which use CC-BY) go further than you do,
so "we have done the best we can" sounds like an apology that "the
best you can" wasn't up to the level that those other, much more
significant projects have managed to do. Should you expect a response
other than "oh well, nice try"?

One of my desires as a free software user, for example, is to be able
to buy a new computer with a complete suite of software preinstalled
on the hard disk, including all the needed apps, development tools,
and documentation and source code for everything, and the freedom to
propagate it all in the same way. All the GNU/Linux stuff and all the
Python.org stuff, plus more general reference works like Wikipedia,
educational materials like Wikibooks, and scientific journals like
PLOS, are licensed in ways that would permit including it with such a
computer. It even includes some entertainment media like various
music downloads from Jamendo and the non-NC CC movies. But your
journal would have to be omitted.

We considered releasing under the GPL, but felt that we wanted to
preserve two things which don't seem to be provided by it:
* Rights of the author to attribution as may be expected and desired
of an academic publication. The GPL doesn't seem appropriate for
disseminating the work of a single author.
* Rights of the author to have their words presented

I would have suggested the GNU Free Documentation License (GFDL) that
Wikipedia uses, but that's just me.

* Opportunity for the author to commercially license their works to
other vendors. By choosing the Share Alike restriction, we have
encouraged the free dissemination of research information without
affecting its commercial use.

Right, the effect on commercial use comes from the NC restriction,
not the SA restriction.

* Rights of the author to have their words presented

I'm not sure what you're getting at but if you mean no modified
versions, you need -ND for that.

It seemed to be the best middle ground between taking a strong
ideological position on either side that would be bound to put
people off side. It preserves some rights for the author while still
allowing a substantial amount of free re-use.

"Middle grounds" often combine the disadvantages of both "endpoints".
And failing to take a strong position can leave you in a weak one.

* Reputation as an
unbiased, financially disinterested group.

The NC licenses withhold some rights for exclusive use by the initial
publisher, releasing only a subset to users. That is usually not a
sign of disinterest. For example, Cory Doctorow's novels are
published under NC licenses, which he says is working for him as a
marketing tool. But he has a clear and undisguised financial interest
in choosing the NC license.

By distributing under the
license we chose, we hoped to establish our credentials.

Well, it seems to me that you're putting out yet another non-libre
publication, it's nice that it's gratis but I don't think it helps
establish credentials with FOSS users and the confusion so far may
actually be hurting.

Your email indicates a possible concern that we are doing something
untoward -- this was not at all intended, nor is it true.

As mentioned in another post, I feel that you're (perhaps
unintentionally) trying to attach to your publication the good
associations created by the licensing practices of the FOSS movement,
while not actually following those practices. "Untoward" might be a
slightly overstrong term, however, you are creating confusion and
maybe suffering from it yourself.

Here is the issue: people who write FOSS code and documentation often
make sacrifices (of their free time, or of potential revenue) in order
to do so. They are aware of those sacrifices and make them anyway.
So it's a bit grating to those of us who work on FOSS projects (or
even just admire them) when someone comes along and tries (even
unintentionally) to attach to themselves the recognition comes from
those sacrifices, without actually making the same sacrifices. That
may explain the reaction you're getting.

 

[Jerry Hill]

And how do you think this is different from any other publication? That
Python Papers is under a CC licence is a red-herring.

Well, the CC license is viral. According to the CC explanation of the
Attribution-NonCommercial-ShareAlike license I am free to create a
derivative work, but only if I then release that work under an
identical license. When I look at the Python Cookbook, it doesn't
seem to encumber derived works the same way. Instead, the Python
Cookbook page says "Except where otherwise noted, recipes in the
Python Cookbook are published under the Python license." The Python
license is extraordinarily broad in what I'm allowed to do with it,
including reusing code in a commercial project.

Again, how is this different from any other publication? Unless you only
read public domain publications, anything you read is copyrighted, and
your arguments apply just as much -- perhaps more so -- depending on the
licence of that publication.

Yes, of course. Thus, I have to avoid using code out of any
publication with a viral non-commercial license if there's any chance
my code will be sold commercially. For the same reason, I won't take
code from a GPL'd product and reuse it in a commercial project without
getting a commercial license from the copyright holder.

By the way, I wasn't really trying to complain about the Python
Paper's choice of license. I just wanted to give my perspective on
why the licensing terms make it unsuitable for me. The copyright
holders are welcome to release their work under any terms they are
comfortable with.

 

[Fuzzyman]

has any of the format zealots posting to this thread actually
volunteered to produce any material for your publication? if not, I
suggest ignoring them. any bozo with a keyboard can contribute stop
energy to a project, but at the end, it's always the people *doing*
things that matters.

+1

You guys are doing a great job and the CC license is fine for free
works like the Python Journal

Fuzzyman
http://www.voidspace.org.uk/python/index.shtml

 

[Fuzzyman]

Maurice LING wrote:
[snip..]

As Steven mentioned -- anything you can read is copyrighted. The
difference is whether is the copyright effective or enforceable. What do
I mean by this? Without copyright, there will not be plagarism. Ask
yourself this question, can you copy William Shakespeare's MacBeth and
submit it as a literary work for a Master of Literary Arts degree? I
believe the candidate will be expelled from university. William
Shakespeare's MacBeth is still copyrighted work but not "enforceable"
because it is pre-1900's work and the author had been dead for more than
50 years. Similarly, works in public domain are still copyrighted --
academically, using work in public domain without attribution (giving
credits in the form of citations) is still plagarism.

This is very, very incorrect.

You're mixing your own moral ethics with an incorrect understanding of
what copyright means...

Fuzzyman
http://www.voidspace.org.uk/index2.shtml

 

[Carl Banks]

Although you might not have intended it, I feel it is still true. You
are misappropriating terminology ("free as in freedom") from the free
software movement that means a work is licensed in a way that meets
some specific criteria, that the license you're using does not meet.

"Free as in freedom" is not terminology; it's a way to differentiate
one of two different senses of the word "free". "Specific criteria" is
some people's idea of what freedom is, but they're not the last word on
it.


Carl Banks

 

[Paul Boddie]

MontyLingua might appear to be GPL-licensed, but then the author puts
some kind of non-commercial clause on top, either thinking that's what
the GPL is all about (out of confusion, perhaps), or believing that he
can limit the rights of the those who license his code. The effect is
that the code is not actually GPL-licensed at all, given the stated
conditions on distribution. See here for more discussion:

http://lists.debian.org/debian-legal/2004/08/msg00622.html

Paul

P.S. This obsession in academia for non-commercial licences is quite a
destructive force: everyone "protects" their own work yet steers clear
of other people's work, redoing what they need themselves, mostly
because they've got an eye on "monetizing" that work later on.
Consequently, the cult of the almighty dollar gets more out of it than
scientific progress does.

 

[Noah Slater]

Thanks for the comments. PDF is, to some extent, a requirement. To
preserve the entire journal as a single "entity" with a reasonably high
production quality, there seems to be no way around it. I could not
find a sufficiently simple way to do multi-format publishing with an
attractive layout and good typesetting.

Which is more important to the Python comunity...

Good typesetting or good, searchable, copyable, usable information?

 

[Fredrik Lundh]

Which is more important to the Python comunity...

the community definitely don't need more random usenet posters who's
only contribution is to complain whenever someone tries to do some-
thing. this thread is an embarrassment for the Python community; you
should all be ashamed of yourself.

</F>

 

[Steven D'Aprano]

Which is more important to the Python comunity...

Good typesetting or good, searchable, copyable, usable information?

Such a sweeping generalization, as if "the Python community" only wanted
one thing.

In any case:

(1) Whether the Python Papers is "good" depends on what content it
includes, not the file format it is published in.

(2) Google can index PDF files. If the Python Papers aren't reachable by
Google's spiders, well, I consider that the Kiss of Death for them.

(3) Code in the Python Papers are certainly legally copyable, and if your
PDF viewer doesn't let you copy text from PDFs, use another PDF viewer:

[quote from page 4 of the Python Papers]
def process(self):
'''
Processes the payment into the financial system
'''
self._recordPayment()
self._despatchOrder()
self._sendConfirmation()
self._concludePayment()
def _recordPayment():
'''
Make a permanent record of the payment in the database
'''
pass

Works fine for me, although I would question the choice of TWO spaces for
indents.

(4) The Python Papers are certainly legally usable, for non-commercial
use. For commercial use, you have at least two legal options:

- treat the work the same as you would any piece of code with an
unacceptable licence, and develop your own NON-derivative version. It's
only copyrighted, not patented: you are still allowed to write your own
independent version. Merely having read a work doesn't necessarily
contaminate whatever you write next.

- contact the author of the work and ask for a commercial licence. You may
have to pay a licence fee, or perhaps not.

Instead of bitching and moaning about it, how about you submit your own
work to the Python Papers under a dual licence? It can be both CC
Non-commercial, and GPL (or whatever licence you like).

I assume the purpose of the Non-commercial licence is to prevent somebody
taking up the Python Papers in full and charging for it. (Personally, I
think that is a very low risk -- who cares if somebody tries to charge
for it? They can't compete with the gratis version being given away for
nothing.) Dual licences for the contributed articles will still protect
the over-all integrity of the magazine from being sold for profit, while
still allowing coders to duplicate useful code and use it in whatever they
choose.

That way, the Python Papers themselves continue to be free as in beer, and
individual articles can, according to their authors wishes, be *at least*
gratis and possibly free as in speech as well.

 

[Guest]

Fredrik Lundh:

Tennessee Leeuwenburg wrote:



has any of the format zealots posting to this thread actually
volunteered to produce any material for your publication? if not, I
suggest ignoring them. any bozo with a keyboard can contribute stop
energy to a project, but at the end, it's always the people *doing*
things that matters.

</F>

+1

The Python Papers look very, very promising! I'm looking forward to
reading next editions. Don't let whiners slow you down. PDF is a good
choice, I personally like to print out longer things I read. IANAL, so I
won't speak for or against CC licenses. They're used widely nowadays for
a variety of applications (music, art, software, publications) so it
seems they're doing their job.

Keep up the good work!

Best regards,
Åukasz Langa

 

[Noah Slater]

I do not think this thread is an embarrassment to the community. I
think it speaks volumes about people's commitment to free software.

While we can applaud such contributions it is no excuse to waiver on
one's ethics and principles.

Regardless of content, or even format, if the Python Papers are not
free as per the FSF's definition they are encumbered.

I am hoping that the author/publishers will reconsider the choice of
licencing.

As an aside, I keep seeing people refer to the CC licences as if they
were all comparable with each other.

This is incorrect. Each licence is very different and saying something
is licenced with a CC licence says absolutely nothing.

http://www.fsf.org/blogs/rms/entry-20050920.html
http://www.linuxp2p.com/forums/viewtopic.php?p=10771

 

[Ben Finney]

I do not think this thread is an embarrassment to the community. I
think it speaks volumes about people's commitment to free software.

While we can applaud such contributions it is no excuse to waiver on
one's ethics and principles.

Yes, this was also my motivation for discussing it here.

Regardless of content, or even format, if the Python Papers are not
free as per the FSF's definition they are encumbered.

I am hoping that the author/publishers will reconsider the choice of
licencing.

I also hope that a free license can be chosen. In the meantime though,
I'm very happy that (as announced in a new thread) the existing
license terms are no longer being falsely described as free.