This document is a partial list of the Roman provincial coins in the ANS's collection for which images are available. Its current purpose is to serve as a test bed for the development of an infrastructure for on-line publication of the society's coins and other objects. The format of this document is a simple list of accession numbers organized by region, mint, and ruler or issuer. Each accession numbers is a link to the society's on-line collection database which contains fuller information for each object. As the system grows in capabilities, more of the information from the database will be formatted for presentation here. It is important to note that most of the records in the society's collection management database have not been individually edited and may be incomplete or contain some errors.

The ANS' electronic publication infrastructure is based on open standards so as to ensure platform independence and the long-term accessibility of the archival files. In particular, all documents are encoded using the Text Encoding Initiative's teixlite XML DTD. At the time of writing, the underlying xml file for this document is available at http://www.numismatics.org/publications/rpcans/rpcans.xml.

The specific capability that this document demonstrates is the rendering of isolated accession numbers in an xml document into text that resembles a numismatic catalog. Inspection of the xml file rpcans.xml will show that it does not contain any descriptive text for each coin it lists. Rather, the tei element 'xptr' is used to mark accession numbers and to indicate that they refer to records in the ANS database. This xml file can easily be translated into html that can then be shown in a browser for easy reading. Similtaneously with this translation, descriptive information for each accession number is read from the ANS's database and this allows a passable numismatic catalog to be produced. The specific advantage of this system is that corrections can be made directly in the database and these will then automatically appear in this document the next time it is translated into html. The translation process is currently implemented in perl in combination with James Clarke's nsgmls SGML processor, which is available at www.jclark.com.

The authorship of this document should be considered institutional as much as individual. Data entry of the Society's collection has been overseen by the curatorial staff and much of it was carried out by volunteers. The current focus is on establishing the organization and scope of the project but attention is also being given to bringing every record up to a consistent standard.