In recent posts Request for CODATA definition of Open Access– and http://wwmm.ch.cam.ac.uk/blogs/murrayrust/?p=445 I was concerned about the use of “open access” to describe a pay-to-access database. I had a very useful and constructive reply from softCon about the “open access” database of spectra which I append below, together with my reply. I will comment in a later post…
PMR: Many thanks for your prompt, full and constructive reply. I hope I can make a similarly constructive response. I will embed comments in your text, which is otherwise verbatim.
Dear Dr. Murray-Rust,
thank you for your e-mail concerning the “UV/Vis+ Spectra Data Base”. Please
let me begin my answer with some additional information.
We started the database in August 2000 and in the beginning all data
(spectra and datasheets) are completely free accessible for everyone. We
thought that this would be helpful to convince the users of the database to
help us in maintaining the database and to convince commercial users
(unilever, bayer, basf, pfizer etc) which benefits from the database or
governmental organizations to provide us with financial support, but that
was naive from our side. During the first six months when the database was
on-line we’ve several thousands of users (commercial and non-commercial) but
we have only 2 (TWO!) users which were willing to help us in maintaining the
database by the provision of data and we’ve got no financial support. Due to
this experience, we’ve decided to change our database policy. The database
was subdivided into a complete free-of-charge “Literarure-Service”
(meta-data) and a “Spectra-Service” (spectral data) for which a subscription
is required. You can access the “Spectra-Service” either by supporting us in
maintaining the database (provision of spectra data) or by paying a moderate
annual fee. Currently almost three fourths of the “spectra-service” users
have complete free-of-charge” access. We make no profit with the database.
PMR: I understand and appreciate the business model. I don’t have any concern about charging for data per se.
To maintain a fast growing database is not only a really hard and never
ending work but also cost-intensively. However, to operate and maintain such
a database financial support is required. Both database services are
operated in accordance to the “Open Access” definitions and regulations of
the CSPR Assessment Panel on Scientific Data and Information (International
Council for Science, 2004, ICSU Report of the CSPR Assessment Panel on Data
and Information; ISBN 0-930357-60-4). We’ve added a link to the original
document on our web-site.
Here are some quotations from the ICSU report:
“…Full and open access” to data implies equitable,
non-discriminatory access to all data that are of
value for science. It does not necessarily equate to
immediate access or ‘free of cost’ at the point of
delivery, although this is certainly the ideal in many
situations, particularly with regard to publicly
funded data. Data should be made available with
minimal delay but a short ‘privileged access’ period
for original data producers may be justified in some
situations. Excessive charging for data that is by
definition discriminatory against some scientists is
clearly contrary to the principle of full and open
access but some cost-recovery is not necessarily
excluded…”
“…There are several economic models for providing
scientists with access to data for research and education.
They include, among others, (1) free and open access to
research data by scientists, with financial support for data
dissemination and preservation assumed by others,
including government science agencies and private
foundations; (2) open access to scientific data for research
and education for the cost of reproduction (that is,
recovering the operational costs of data dissemination);
(3) free and open access to metadata, and cost-recovery
pricing for data (or data licenses) in order to support the
full data infrastructure. When this last approach is
employed by a commercial company, the financial charges
for data must be sufficient to recover all investment costs
and to make a profit for investors. An important variation
on this includes licensing for scientists to use specific
bodies of data at reduced cost…”
Thank you very much for this. I may comment later in a blog that it is unfortunately for the “open access” publishing community that ICSU has chosen the phrase “open access” to mean an affordable charge structure”. I accept that by some standards 100EUR is non-discriminatory.
Indeed we are one step ahead to the ICSU recommendations since we provide
free-of-charge access to the meta data/related data without any
cost-recovery and in addition the database user can decide if he is willing
to help us in maintaining the database or to pay a moderate utilization fee
which ensures that the database will be operated, developed and maintained
in the future.
Again, currently all meta-data (datasheets) are free accessible as well as
other related data (e.g. software, satellite-data etc t.b.d.).
Finally, as mentioned in the ICSU report “WHO PAYS – Data production and
management are costly”. We’ve currently no idea how to finance this database
except by charging some of its users with a moderate fee. Do you have any
ideas?
PMR: I agree that data are costly, though technology brings some costs down. With Open Access in the publishing sense there is a strong movement towards author-pays supported-by-funder. The major charities (Wellcome, HHMI) are making allowances for authors to pay for publication as Open Access (toll-free access and hopefully re-use).
I suggest you have a look at what the NIST group (Michael Frenkel and colleagues) have done with ThermoML. Here the publishers have a model where if thermochemistry is published (there are 4 or 5 journals) it has to be in ThermoML and has to go into an Open Access database. This seems to work to everyone’s benefit. I’ll write more later … but this might be a useful model for you. It won’t pay YOU directly but may create a data stream at near zero cost.
PMR: more comments in later post…
Peter, I hope that these information will give you an idea about our
intensions.
Best regards,
Andreas
————————–
Dr. Andreas Noelle
science-softCon
Auf der Burg 4
63477 Maintal
Germany
Phone: +49 6181 498414
Fax: +49 6181 498415
e-mail: andreas.noelle@science-softcon.de
www.s-sc.de
VVV