[WikiEN-l] [Foundation-l] CAS Discourages Using SciFinder to Help Curate Wikipedia

[phoebe ayers]

On Sat, Mar 8, 2008 at 8:13 AM, David Gerard <dgerard at gmail.com> wrote:
> http://wwmm.ch.cam.ac.uk/blogs/murrayrust/?p=997
>
>  They have a specific hate-on for Wikipedia:
>
>  "Chemical Abstracts Service (CAS) objects to anyone encouraging the
>  use of SciFinder - and STN - to curate third-party databases or
>  chemical substance collections, including the one found in Wikipedia."
>
>  The claim is that "CAS numbers are copyright CAS/ACS who have the
>  legal right to regulate their use - as above." I find this idea highly
>  dubious myself, though I wonder what countries it would legally fly
>  in. Particularly given that "CAS identifiers have come to be accepted
>  as a primary identifier system for chemistry."
>
>  Anyone here have informed legal commentary?

I don't have informed legal commentary, but my guess is -- and their
actual comment bears this out -- that ACS's objections are due to the
extremely stringent license requirements that one agrees to for when
one subscribes to SciFinder. Basically, redistributing information
from SciFinder/CAS Abstracts to other people who didn't subscribe
(i.e. via Wikipedia, or otherwise) is almost certainly not kosher. I
make this guess based on the rules that libraries agree to when they
subscribe to SciFinder. There are nearly always specific provisions in
the contract (not just for SciFinder, but for most databases and
journals) that say who can get access to the data -- for instance,
only the faculty, staff and students of a university. Obviously
enough, the many glorious readers of Wikipedia are unlikely to fall
into this faculty/staff/student classification.

Assuming this is true for most contracts they sign, then probably
anyone systematically posting data from one of these systems is
violating some provision of their contract. This doesn't have anything
to do with copyright law per se, but it is a terms of use question.

The separate issue of CAS numbers being copyrighted is very
unfortunate, but everyone is right that the CAS numbering and indexing
and data organization scheme that is the core of SciFinder and related
tools is very much the lifeblood of ACS, and they are afraid of losing
it. Anecdotally, though the ACS is a nonprofit, they have historically
not been inclined at all to go to open access/open licensing/open
content schemes, unlike some other major publishers who are starting
to see the wisdom of making information freely available. So I'm not
sure that an appeal to their better nature would be terribly likely to
work. But, it's always worth trying.

-- phoebe