Hi,
at the University Library of Cologne (www.ub.uni-koeln.de, Germany) we
use the open source software OpenBib (www.openbib.org, GPL) as our
central search portal named KUG (Koelner UniversitaetsGesamtkatalog =
Collective catalogue of the University of Cologne,
kug.ub.uni-koeln.de) for the library catalogues of the numerous
institutes (around 145 in 110 separate catalogues) at our university
as well as 'other interesting sources' like local OAI repositories,
special collections etc.
At the moment KUG consists of 6892554 titles in 139 separate
catalogue databases.
Although we seem to live on a different planet in germany
library-wise, with other data formats (MAB instead of MARC21) and
cataloguing rules (RAK instead of AACR2) our KUG might still be
interesting for a short announcement to this list.
The main principle behind KUG is to build a central system with
separate uniform databases that is fed daily with the data of
different sources, mainly library catalogues. In this way we can do
stuff like enriching our catalogue data (e.g. ISBN10/13) while loading
into KUG, nothing new after all. Because we end up with *uniform* data
in our KUG, several additional features across all our catalogues are
possible. Time critical availability and holding information is merged
from the corresponding separate ILS on-the-fly with web-services.
OpenBib is a typical LAMP application using (Debian)Linux, Apache with
mod_perl, MySQL, Perl, Xapian (www.xapian.org) as our search engine
and the Perl Template Toolkit for flexible templating. Currently
version 2.1 of OpenBib consists of:
- 57 Perl-modules (18140 LOC) + 38 programs (7276 LOC)
- 377 templates (154 default)
- 5 system-databases: config, enrichmnt, session, statistics, user
- 7 data-backends: Aleph, Amarok ;-), FileMaker, generic, Lidos, OAI,
OCLC SunRise
We use jQuery for JavaScript and AJAX, but every KUG feature is also
available without JavaScript enabled - in this case you'll have to
click one more time, but that should be all.
The *productive* version of KUG is always available from OpenBib's CVS
repository at BerliOS:
http://developer.berlios.de/projects/openbib
Some features of OpenBib and KUG are:
- RSS-feeds for new titles in each catalogue. Additionally we offer
RSS-feeds for new titles based on a given person, corporate body,
classification and subject heading.
- Facetted search with drilldowns using the Xapian search engine in
our simple search - complex search is still 'pure SQL'.
- Tagging
- Scores/Reviews (currently not available in KUG)
- Mashup with BibSonomy (www.bibsonomy.org) - from its description 'a
social bookmarking and publication sharing system' - using
bibliographic hash keys for different types of lookups. You can browse
the publications and bookmarks of BibSonomy in KUG based on tags and
get holding information for our local libraries. Additionally you can
automatically sync your local tagging behaviour in KUG to BibSonomy.
So BibSonomy is fed with local tags and library data and is
effectively acting as a central 'social' repository. In our
Wikipedia-Mashup we have integrated (a simple) search for persons and
ISBNs as well as a list of all Wikipedia articles that reference the
current title.
- (Tag)Clouds for different types of data, e.g. most frequently used
persons, corporate bodies, classifications, subject headings, years of
publication, Tags, literature lists and search terms
(Search->Overviews->Clouds)
- Analysis of KUG usage resulting in recommendations of related
titles, sorting of search results based on popularity, top 20 of the
most used titles for each catalogue, etc.
- PermaLinks for titles and (public) literature lists
- Availability information and links to Google Book Search (and
BibSonomy) for a given title.
- Several external 'connectors' to get library data out of our
catalogues, e.g. UnAPI (http://unapi.info/) for use in Zotero
(www.zotero.org). Supported formats are currently BibTeX, MODS,
OAI_DC.
- Central enrichment database for additional info across all
catalogues based on ISBN or the bibliographic hash key (e.g. for
tables of contents, locally available e-books) Based on this
enrichment database we also implement 'virtual systematic browsing' -
so although a given catalogue doesn't know anything about the
classification of a given title this title can still be 'browsed'
through the central classification data.
- Other manifestations of a given work (using thingISBN)
- Flexible use of our templating system, so it is possible with
minimum effort to create different 'views' on our
catalogues. Currently we offer around 144 different views
(Search->Overviews->Views) ranging from simple views with preselected
catalogues to visually separate search portals like
http://portraitsammlung.ub.uni-koeln.de/
http://einbandsammlung.ub.uni-koeln.de/
http://landschaftsbilder.ub.uni-koeln.de/
http://richterbibliothek.ub.uni-koeln.de/
- Gathering of statistical data with an event-based statistics
framework
- A web based administration interface as our control center
- 'Intelligent' load balancing using load averages on each server.
- and of course UTF8, I18N etc.
A few of these features are scheduled for the next release in mid/end
of June.
Unfortunatly our currently used KUG version 2.1 isn't localized to any
other language than german, but an experimental localization to the
english language (excluding help texts) is available on our
development system with the upcoming version 2.2 of KUG/OpenBib. On
this system you can also have a look at all of the features (with
hopefully few bugs ;-). Apart from bug fixing the status of this
system is 'frozen':
http://kug5.ub.uni-koeln.de/
To view some of these additional features I recommend this title in
the catalogue of our institute of applied computer science (in its own
'view'):
http://kug5.ub.uni-koeln.de/portal/connector/permalink/inst006/6439/1/inst006/index.html
More information on OpenBib and KUG (mainly german) can be found on
these pages:
http://www.ub.uni-koeln.de/dezkat/content/veroeff/index_ger.html
http://www.ub.uni-koeln.de/dezkat/content/stimmen/index_ger.html
Currently here in germany quite a lot of NCG-projects appear from
different libraries as opposed to big commercial vendors. These
include
- E-Lib (http://suche3.suub.uni-bremen.de/)
- XOPAC (http://www.xopac.org/)
- HEIDI (http://www.ub.uni-heidelberg.de/Englisch/helios/kataloge/heidi.html)
- beluga project (http://beluga.sub.uni-hamburg.de/blog/)
- vibi (http://is.uni-sb.de/vibi/suchen.html)
Regards,
Oliver
--
Universitaet zu Koeln :: Universitaets- und Stadtbibliothek
IT-Dienste :: Abteilung Universitaetsgesamtkatalog
Universitaetsstr. 33 :: D-50931 Koeln
Tel.: +49 221 470-3330 :: Fax: +49 221 470-5166
flimm_at_ub.uni-koeln.de :: www.ub.uni-koeln.de
Received on Tue Jun 03 2008 - 03:26:21 EDT