Arachne is the central object-database of the German Archaeological Institute (DAI). In 2004 the DAI joined the Research Archive for Ancient Sculpture at the University of Cologne (FA) in providing and further developing Arachne as a tool for free internet-based research. Arachne's database-design uses a world-model that tries to build on one of the most basic assumptions one can make about archaeology, classical archaeology or art history: all activities in these areas can most generally be described as contextualising objects. So Arachne tried to avoid basic mistakes of earlier databases that limited their object-modeling to specific project-oriented aspects, thus creating separated containers of only a small number of objects. All objects inside of Arachne share a general part of their object-model, to which a more class-specific part is added to describe the specialised properties of a category of material like architecture or topography. Seen on the level of the general part, a powerful pool of material can be used for general information-retrieval, whereas on the level of category-properties, very specific structures can be displayed.

Started in 1995 as a database for ancient sculpture on FileMaker Pro, Arachne continued to develop, and, after 2001, benefited from the newly established chair for Humanities Computing at Cologne university. Since then Arachne has also been a testbed for students engaged in a complex and serious development project.

Thanks to significant and ongoing support by the Deutsche Forschungsgemeinschaft since 2001, Arachne started to integrate negative-archives of ancient sculpture that went beyond the specialised documentation retained in Cologne itself: the Malter- and Fittschen-archives, and the negatives of ancient sculpture of the German Archaeological Institute in Rome beginning in 2003. All that totaled 40,000 high-qualitiy-scans of ancient sculptures that are presented with state-of-the-art scientific documentation. Starting in 2006, the digitizing of historic-glass-negative-collections will result in another 40,000 digital images, beginning with those of the German Archaeological Institutes in Athens, Cairo, and Istanbul. Beside this larger project, many activities are ongoing on different levels, for example the online preparations for the "Corpus der Antiken Sarkophagreliefs."

Since 2004 Arachne has been completely reworking an building the system from scratch using an "AMP" (Apache/MySQL/PHP) database-management-system with pure web-standard technologies. Based on technical foundations that seem to be viable for the next decade and being strategically positioned as a central object-database for a large institution that counts about 2 million images inside their photographic archives and produces even more data each year in the course of its research activities, Arachne's potential is not modest. To meet it, there is still a long way to go.

In week 11 of 2006, Arachne serves about 1,146 registered users who can access 116,000 scans and 96,000 objects free of charge. Quality of data differs regarding the state of documentation and has to be improved in the areas where just a basic description is available. Forecasting the target area in quantities, one could refer to statistical estimations of huge digitization projects of cultural heritage in Europe. They foresee being able to digitize approximately 30% of the existing photographic material under the current financial circumstances. That would yield in the region of 700,000 digital images from archive material. With the addition of newly produced documentation, one could hope for 1 million 20 or 30 years from now.

All objects should then reflect their contextualisations in basic categories of material and topography. All categories of material cultural heritage from European antiquity as well as from Asia and South America should then be integrated. Regarding interoperability, different steps have been taken to ensure the widest possible use and visibility of data, while at the same time acknowledging the liabilities of intellectual property-rights management and copyright protection that DAI and FA accepted when documenting the framework of collections, museums, excavations etc.

Technical aspects of interoperability do impact the way data are shared and connected inside DAI itself, mainly between Arachne and several GIS-systems that are used on DAI-excavations and surveys -- so a given object will only reside once inside the overall dataspace of DAI. In that sense DAI will support uniform resource names in the near future to unmistakable identify objects residing in Arachne. It is the policy of DAI and FA to implement the CIDOC Content Reference Model, for which different project-applications are ongoing. Also in preparation is the support for the Open Archives Initiative. As a partner of CLAROSnet, Arachne is committed to work on the issues mentioned above as well as to build a multilingual interface that it lacks today. It was the policy of DAI and FA to address issues of that complexity not on their own but in an international framework. A solution must be found for this in the near future.

On the technical roadmap for the coming years is finalizing the integration of specific material categories. Therefore, the Arachne thesaurus will have to be expanded. A better treatment of search results will have to be implemented, as well as a clearer history function. The general problems in displaying huge search-results are well known on the web or in databases as such. Arachne is no exception here and will have to integrate its "context browsing" feature with a topic-map design, that allows switching between more graphical and more textual displays of information.

To learn more about Arachne, please see http://www.arachne.uni-koeln.de/ .

-- author