[CDAO-discuss] terms of reuse

[Hilmar L. <hl...@ne...>]

Hi all,

I saw that the license attached to CDAO on the SourceForge project  
page is the GPL. This is indeed also what's stated in the paper, so I  
should have noted long ago.

The GPL is, I think, a rather bad choice for licensing CDAO. First, it  
was designed for software, and how to apply the terms to an ontology  
are not so clear. It's specifically not clear how to interpret the  
"viral" terms of the GPL in the case of someone reusing an ontology,  
provided that such viral terms were actually desired. Second, as I  
think we are all interested in the broadest possible reuse of CDAO, a  
"share-alike" requirement in reality I think does more harm than good.  
Third, as a license GPL actually implies (and asserts) copyright.  
While in some ways one may see an ontology as a work of creative art,  
and hence as eligible for copyright protection, much of the things  
asserted in CDAO (as in most other life science ontologies) are facts,  
which at least in the US are not eligible for copyright protection. So  
at least it's a grey zone situation. Perhaps more importantly,  
copyright protection and licensing are *legal* instruments, useful for  
enforcing one's rights in a court of law. However, what we are  
probably rather interested in instead is that certain community norms  
for giving credit (attribution) are met by those reusing the ontology  
wherever this can be reasonably expected. A public-domain dedication  
can state those expectations just as well, and possibly better, than a  
license, while removing all barriers to reuse. (For example, licensed  
content with copyright assertion, whether with solid or questionable  
legal standing, creates all kinds of headaches in aggregations that  
would result in mixed licensing situations.)

Therefore, I would suggest that the CDAO license be changed to a  
public-domain waiver, with an attached declaration of expectations for  
attribution.

As an example for how this could look like, here are the terms of  
reuse for the Teleost Anatomy Ontology (TAO), as contained in the OWL  
file [1]:

"To the extent possible under law, Wasila M. Dahdul has waived all  
copyright and related or neighboring rights to the Teleost Anatomy  
Ontology (TAO), under CC0 license (http://creativecommons.org/publicdomain/zero/1.0/ 
). This work is published from the United States. It is requested that  
users of this vocabulary cite the following publication: Dahdul, W.M.,  
Lundberg, J.G., Midford, P.E., Balhoff, J.P., Lapp, H., Vision, T.J.,  
Haendel, M.A., Westerfield, M., and Mabee, P.M. (2010). The Teleost  
Anatomy Ontology: anatomical representation for the genomics age.  
Systematic Biology 59, 369-383. http://dx.doi.org/10.1093/sysbio/syq013"

(Note: In comparison, the CDAO at present actually has no such  
statement in the OWL file, so a machine accessing the ontology would  
have no way to know. [And yes, it would indeed be pretty hard for a  
machine to extract the above as the pertinent terms of reuse from the  
OWL translation, as I've explained on the OBO format list [2]. But  
that's because it is maintained in OBO format - for CDAO we could  
choose how to expose this directly in OWL.])

	-hilmar

[1] http://purl.obolibrary.org/obo/tao.owl
[2] http://bit.ly/uZBlae

-- 
===========================================================
: Hilmar Lapp  -:- Durham, NC -:- informatics.nescent.org :
===========================================================