Development of Creative Commons licenses

[Nancy Ide <ide AT cs.vassar.edu>]

> I either don't understand what you are doing, or how the CC  
> licenses could
> cause you a problem.

Apparently I have asked my question to the wrong group.

> Are you talking all of those unmarked, but by default "all rights  
> reserved"
> copyright texts out there in the wild?

No, I am talking about texts on the web that are explicitly indicated  
to be reusable by others--mainly, these days, using CC licenses. All  
of those "all rights reserved" texts are completely out of the  
question for our purposes, as we cannot violate copyright laws since  
we re-distribute the texts.

We love the CC "By" license, but the problem is that the CC makes it  
pretty clear that the "By/Attribute/non-commercial" license is the  
"right" one to use, and therefore very few authors use "By". And for  
the purposes it was intended to serve, "By/Attribute/non-commercial"  
may well be the right license. I am trying to make it clear that  
there is another purpose that the CC licenses do not address.

> If you can use those withoug contacting the copyright holders for  
> permission,
> as a fair use, I would think you could apply the same fair use no  
> matter what
> the license. (I am not a lawyer and that is a guess.)

Bad guess. Fair use, although not technically defined, is understood  
to cover no more than about 250 words.

Now, lots of people in our business do use web texts, all rights  
reserved or not, as a way to mine language data. Since they are not  
reproducing or redistributing the texts but just gathering statistics  
from them, they will never have a problem (this is what companies  
like Google, Microsoft etc. who have plenty of money to develop a  
corpus on their own do). In our case, we want a corpus we can add  
linguistic information to (like, the part of speech of each word,  
syntactic structure, etc.), is in American English, and is balanced  
for genre (that means different text types, like fiction vs.  
newspaper, etc.)--which web data randomly acquired definitely is not.  
And above all, we need to distribute it to other linguists who can  
use the information we have added to the data.

We are actually trying to make the data as freely available as  
possible without infringing on anyone's rights. As I said, we have  
had several authors give permission to include their texts in the  
ANC--it is however impossible for us to contact every web author and  
ask for such permission. We are very much in favor of copyleft, and  
distribute all our software and other data via GNU licenses that way.  
We are primarily researchers and scholars, with a few publishing  
houses (mostly university presses) thrown in--who have, by the way,  
contributed lots of texts already but do not have genres like blogs  
and email--and we are all for sharing resources via copyleft as this  
is very much in the spirit of academic research. So once again, for  
those of you who didn't get it the first time, we do not have an  
issue with share-alike for the purposes CC authors use it now.

Our issue with CC is simply, why couldn't there be a license that  
allows for research use, along the lines I described? I.e., ability  
to re-distribute the text, with full restrictions on "re-publishing"  
all or part of it (except within fair use) in any other form except  
as a part of a research resource, and to use it for linguistics  
research (which may or may not--most of the time, not--result in a  
commercial by-product)?

> If you were expecting the commons community to be just a bunch of
> work "free for exploitation", then you have a serious misunderstanding
> of the community.

I was hoping (not expecting) that the commons community would be open  
to another type of use that they do not currently address. I see,  
however, that you are not, and furthermore that you in turn have a  
serious misunderstanding of the linguistics and computational  
linguistics communities and their research. That's not unusual, but  
you'll be happy one day when your search engine actually returns  
meaningful results.

Nancy Ide

=======================================================

Nancy Ide, Ph.D.

Professor  and Chair of Computer Science
Vassar College
Poughkeepsie, NY 12604-0520 USA
Tel: +1 845 437-5988 Fax: +1 845 437-7498
ide AT cs.vassar.edu

Chercheur Associe
Equipe Langue et Dialogue, LORIA/CNRS
Campus Scientifique - BP 239
54506 Vandoeuvre-les-Nancy FRANCE
Tel: +33 (0)3 83 59 20 47 Fax: +33 (0)3 83 41 30 79
ide AT loria.fr

=======================================================