BAM, the joint portal for Libraries, Archives and Museums in Germany, considers itself to be a digital memory institution. Currently the portal holds more than 40 million records from a wide range of cultural institutions, some 37 million data sets from six libraries or union catalogs, 2.9 million data sets from eleven archives, 300.000 data sets from twenty museums and 800.000 data sets from other institutions.
These significant differences in numbers of data sets are not only due to the size of the holdings of the participating institutions but also to “cultural differences” between libraries, archives, and museums in creating records and collaborating in union catalogs.
The paper describes those differences from the perspective of the BSZ, the hosting organization of BAM, and a major contributor to BAM, the Foundation Prussian Cultural Heritage (Stiftung Preußischer Kulturbesitz), Berlin. The point of view is specific for the situation in Germany and might differ from the situation in other countries. There are certainly other important issues that are not mentioned here as we chose to take a perspective specific for BAM.
Introduction:
Knowledge organization got its name in the context of enhancing the processing of information throughout an institution. In a networked environment knowledge organization should provide for the optimal allocation of information resources to the right person(s), at the right time(s) and places(s), in an expectable and understand-able format. In the past knowledge organization was occupied with the classical topics of controlled vocabularies: classifications, thesauri, their theory, development, and usage. The topics have not changed dramatically, but the growing impact of the internet has shifted the focus somewhat to such topics as metadata standards, ontologies, semantic web etc. The politics and ethics of knowledge acquisition and distribution was and is a main topic, too. Here knowledge organization shows strong ties to the social sciences. The expanding space of internet services has brought together the different language communities, but in most of the cases by neglecting the vernacular language of the internet user. English is most common and seldom perfectly spoken or understood by non-native speakers: so there are modern variants (beside American English e. g. continental English) and dialects (e. g. conference pidgin). Automatic translation produces funny results normally; it works only in very specific environments with a basic vocabulary of around 5 words (e. g. most sciences and applied sciences). So there is a strong need of processing information on multilingual platforms for the cultural heritage (ch) domain. There are many endeavors to meet these needs, but none is convincing until now. Therefore several possible solutions will be discussed below.