[open-bibliography] Tomorrow: 4th Virtual Meeting

Peter Murray-Rust pm286 at cam.ac.uk
Tue Oct 5 10:52:45 BST 2010

On Tue, Oct 5, 2010 at 8:44 AM, Adrian Pohl <pohl at hbz-nrw.de> wrote:

> Oups,
> I am very sorry for that. Obviously, I was quite careless yesterday. (I
> copied and pasted that and then forgot to change the dates.) As it seems,
> you momentarily can't access the okfnpads. As soon as I can I will correct
> that.
> > Is it today (October 4 i.e. about now?) or tomorrow (October 5)?
> The meeting is today (October 5), 16:00 BST.
> I will try to be online - I shall be in the British Library. I will see if
there is somewhere I can use audio.

I have got an important publisher on board who is happy to make all their
bibliography Open. I'd like to see if we can come up with a set of
guidelines - that we agree. It's a bit rushed. I don't want to announce
names till we have agreed.  Here is my suggested starting point - it's based
towards science publishing and online material

individual works (articles, monographs, etc.) are described by a
**bibliographic entry**. This is a necessary and unique collection of
information that allows:
    - addressing (how do I find the work)
    - identification (what is the work that I have found)
If the resources are electronic then the address can be a URL and the
identification a URI (possibly based on one ore more identifier systems)

There are conventional many (semi-standard) metadata fields which describe
the work and are useful for bother discovery and identification. They
* title of work
* authors, possibly including addresses and other contact details
* publisher
* dates
* title and identification of enclosing work (e.g. a journal)
* format of work (language, mimetype, document structure (number of words,
number of images, number of tables), manifest of associated files)
* rights associated with work
* abstract (usually a paragraph describing the contents of the work)
* sponsorship (e.g. funding)

We assert that this information associated with an indivdual work is in the
public domain and is compliant with the Open Knowledge Definition. To
protect its nature we wish to apply a licence asserting this nature, such as
PDDL or CC0. It follows that an indivdiual bibliographic entry derived from
the work itself is free of restrictive rights. This information can be
derived either by copying it from the pysical work itself or by visiting the
manifestation of the work on the Internet.

We agree that collections of bibliographic entries may have property rights
associated with them, but that individual entries derived from the works or
from an Open collection do not carry these restrictions.

We would include some community norms. These are not covered by the licence
but are aspirations that the community holds and which might be invoked to
restrict certain types of usage:
* a re-publisher of bibliographic information should make best-endeavour to
ensure its correctness.
* large-scale robotic collection of information may need to be agreed with
the publisher to avoid load on server
* re-publishers should avoid asserting any non-explicit  endorsement from
the publisher

Note that this assertion about the bibliographic entry for a work makes no
statement about the rights (if any associated with):
* a list of citations or a bibliography in the work
* content of text other than the abstract and fields above
* content of tables
* content of images
* content of associated files

and we believe they should be covered by separate protocols.

There are many benefits to a completely Open scientific bibliography:
* discovery of works is made easier (especially through new technology).
This enhances the visibility of works
* identification of works is improved, leading to fewer mistakes in
recording bibliography.
* tools and resources can be created for scientists to speed up and enhance
quality in authoring
* bibliography can be made part of the Linked Open Data vision for the Web
* new analysis tools can be created for analysing patterns of research and


They are happy for their abstracts to be in the public domain - this is a
major win. If we can get the rubric right and agreed it is a major spearhead
to get other publishers on board. If we can win ones such as nature then I
think most of the rest will follow.


Peter Murray-Rust
Reader in Molecular Informatics
Unilever Centre, Dep. Of Chemistry
University of Cambridge
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-bibliography/attachments/20101005/f75f719f/attachment-0001.htm>

More information about the open-bibliography mailing list