> The question "who really possess the copyright" of such data
> mainly produced at academic research institutions has no
> clear answer.
Not so. I am no expert on copyright laws, but I do believe one can
obtain copyright protection on a collection of information, even if the
contents are in the public domain, provided that the collection itself
is new. Further, one can copyright the way information is formatted.
Finally, corrections, reference numbers or other cataloging information
added by the information compiler is also subject to copyright
protection. I am pretty sure of this for the U.S.A. and suspect this is
true in most of the world.
What does this mean? As a U.S. government employee, my research is not
subject to copyright -- anyone in the world can use my software,
structures,... in any way they choose and I am glad to contribute it.
(Congress specifically exempted reference data from this rule, BTW). On
the other hand, if one goes to the library makes and distributes 10,000
photocopies of a table of coordinates from one of my papers, that is a
violation of the journal's copyright. (Sending one copy to a friend
probably does not violate the copyright, since there is an exemption for
"scholarly purposes".) What if you retype/reformat the coordinates and
then make 10,000 copies? In this case I'm not sure, but since the
crystallographic data are not protected, I suspect this is indeed legal.
What about coordinates from, say, Armel's papers? That depends on the
copyright agreement Armel made with the journal. He may have given
ownership to the journal. Depending on that agreement, perhaps even
Armel will be limited in what he can do with "his own" data.
So, contrary to the assertions made on this mailing list, when you
republish crystallographic data, using a database's format and/or
include information generated by the database compilers, you are
probably violating their copyright. If you reformat the information and
restrict yourself to the data that appeared in the original article,
then things are less clear and probably depend on many factors.
All the legal B.S. aside. I personally would like to see
crystallographic databases made widely and cheaply available. Given that
no government seems to want to pay to give them away, they probably will
never be free. Volunteer efforts, like Armel's powder data repository
(http://sdpd.univ-lemans.fr/powbase/index.html) could change this -- if
lots of folks contribute -- now that gigabytes are cheap and web access
is nearly universal. Such efforts also have the potential to keep
database pricing in line by providing an alternative.
I am no great friend of database "vendors." I have gotten in trouble
for publicly criticizing the representative of a database organization
that charges exorbitant prices to U.S. companies who want occasional
access to their data compilations. But, this is not the case with the
ICSD:
* they offer a very good academic discount;
* they allow pay-as-you-go searches via STN for the person who does not
want to
> pay for 100 000 entries when you need, say, 10 data sets in one year;
* even their commercial price seems pretty reasonable (to me) compared
to some of the other databases.
Microsoft may or may not set prices based on the the number of people
who buy their products, but the ICSD academic discount would not exist,
if not for all the folks willing to pay it. I will not defend Bill --
but please don't rip off the folks who are doing all the right things.
Brian
(The above represent my own opinions and will probably get me in even
more trouble).
