Development of Creative Commons licenses

[Terry Hancock <hancock AT anansispaceworks.com>]

Nancy Ide wrote:
>  To clarify: our use is this: we gather texts, "annotate" them with
>  linguistic information, and distribute them for use by mostly
>  researchers, but in a small number of cases, commercial enterprises,

The "annotated text" is outside of fair use, is a derivative, and the
NC or SA terms will apply to it, IMHO.  (IANAL, etc.).

>  which in turn consist mainly of publishers who are developing either
>  dictionaries or English-as-a-second language (ESL) texts. There are
>  two major uses: (1) use the texts to generate statistics about
>  usage, like, how many times a noun is followed by a verb, etc.

These statistics are "data", are allowed by fair use, and fall outside
of any copyright terms (CC or ARR).

> (2)
>  looking for patterns, like how many times is " could care less" used
>  instead of "could not care less" in current usage. In case (1), the
>  researchers never even look at the text, they just generate
>  statistics from it. In case (2), most of the time a concordance,
>  consisting of a set of sentences, is looked at, say, to find all
>  occurrences of "could" followed within some number of words by
>  "care".

This would appear also to be "data", not "the work" or "a derivative
of the work".  It's fuzzier though: if one work is used, and appears
completely, but scrambled. However, I suspect most courts would
consider that sufficiently "transformative" as to be fair use,
particularly as it is used with text only.  In my opinion, anyway, it's
still fair use.

>  The idea is to see how people use language. So, as an example
>  of a commercial use, a dictionary maker or an author an ESL textbook
>  might point out that "could care less" is more frequent than "could
>  not care less" in current usage. They might, in a rare instance, use
>  a sentence from one of the texts that was shown in the concordance
>  to use one of these constructions in their published text or
>  dictionary.

Using a sentence from a longer work is fair use, of course. Especially
when it is used as a usage example like this. That's one of the most
fundamental and well-defined examples of "fair use": a quote for academic
use (doesn't matter if the academic use is commercial or not).

> > Since you indicate that you can't use "ARR" works under fair use
> > then I would expect you also not to be able to use most CC works
> > under fair use as well, but only under licensed use.
>
>  This is really unclear, and any help you could give is welcome. We
>  are considering that maybe, we could use the "By/Attribution/Non-
>  Commercial" stuff in the ANC, especially if we eliminate a chapter or
>  two from any text we download.

I think you're out of luck with NC, for anything that exceeds fair use.
You definitely are making "commercial use" of the work, according to
the broad guidelines established by CC.

> Consider the scenario outlined
>  above-- is any of this in violation of the default CC license? We
>  *re- distribute* the texts, but we do not *re-produce* the texts, and

Um. The annotated full text is a reproduction.  So that's not quite
true, unless I misunderstood you earlier.

>  by the license any user has to sign with the Linguistic Data
>  Consortium (http://ldc.upenn.edu), which distributes the ANC, no
>  re-production is allowed.

If you are using SA works, those terms are invalid.

You should note that the SA applies only to the derivative of the work,
not to the rest of the stuff on the same disk, so the disk as a whole can
still have your restriction, since restricted work would be copied to.
However, any SA work on the disk could be legally copied under the
SA terms, including your annotations.

> So, what do you think--can we use
>  CC-default -licensed texts or not? I think this is a tricky question,
>  one which I had hoped this forum could answer or at least address.
>  Can you help?

I reiterate, there IS no "CC-default -license".

The CC-By-NC-SA is the most popular license because it is the one
that authors most often choose, not because they "defaulted" to
it.

> > If you could do your whole project under a copyleft plan, I would
> > donate my stuff if you wanted it.
>
>  We have different licenses for different parts of the corpus--our
>  biggest problem being the ARR texts. Easily, your materials can be
>  distributed under copyleft!
>
> > I am a copyleft fan. I can give you lots of my stuff under a BY-SA
> > license if it would help.
>
>  Anything helps! Please look under "Contributing" on our site and send
>  us your stuff--assuming you are a native speaker of American English
>  (see the ANC website for a definition).

So. Are you saying you are willing to change the license to provide
sharealike terms? If so, then I see no problem for CC-By-SA works.
Do it.

I don't think it'll help with NC works, though.

> > Thanks for your replies. I am not so sure the tone was called for
> > wrt my original posting though.

I'm sorry if you found my response "bracing", but I'm not sure
what you can possibly have objected to. I answered your questions,
and I attempted to dispell your apparent misapprehension of
the purposes of CC licenses.

I am inclined to support your research -- but you must realize that
if you are using copylefted public good works to create, then the
outcome should be a copyleft public good as well.  That's just
quid pro quo -- those are OUR terms of use.

Cheers,
Terry

--
Terry Hancock (hancock AT AnansiSpaceworks.com)
Anansi Spaceworks http://www.AnansiSpaceworks.com