Cover Pages: Markup Standards for 'Name and Address' Information.

As XML vocabularies, DTDs, and Schemas continue to proliferate, rising in number from hundreds to thousands, one feature remains consistent: markup models almost always provide some mechanism to represent name and address information for personal and corporate entities central to the application domain. XML markup schemes for name and address sometimes draw upon established standards or data models, but usually do not. For simple applications, a few data elements are adequate to represent the name of a person, corporate entity, or geographic location; for complex applications (e.g., criminal records and health records management, genealogical and prosopographical research, bibliographic name authority), dozens or hundreds of information objects are required to represent a name and its relationships over time. Similarly, simple forms used for collecting address information typically use a dozen data fields; specialized applications (e.g., land and property ownership, international postal mail, GIS-aware database systems) require hundreds of data entities to model the notion of address. Name and address data models are often deeply integrated into personal information database frameworks. Reconciling the different models is not easy because the problem domains are so different: variably-scoped information spaces and conceptual/analytical perspectives unique to different models give rise to fundamentally different factoring of the same or similar information into metamodels for discrete factoids. Models competent to support machine processing of internationally-standardized name/address data must reckon with linguistic variation and differing cultural conventions. Significant standards work now underway will provide the basis for some re-use of markup models across application domains, but we expect convergence to be slow. Ubiquitous computing levies new requirements to model features of 3-D physical location and recipient-authentication along with traditional address data.

An example of heavy requirements for markup representation of name may be drawn from the domain of library automation, where (for example) the Anglo-American Cataloguing Rules has recognized authority. The AACR 2 volume dedicates a full 100 pages (chapters 22-24, pages 379-479) to the topic of personal, corporate, and geographic names, offering guidelines for sub-element decomposition and aggregation adequate to support name authority control in a bibliographic context. A reflex of this may be seen in the MARC 21 Format For Authority Data: Field List and MARC 21 Concise Format for Authority Data.

With the advent of ubiquitous computing there are many new applications for address/location management (and "location" modeling). In HP's Cooltown project, for example, there is tight integration between physical location description (e.g., absolute location expressed in a 3-D coordinate system) and geographic location description based upon zip codes, street names, visual landmarks, service centers, buildings, floors, rooms, etc. As users become increasingly mobile, models for "address" will need to quickly accommodate additional notions of proximity, direction, accessibility/reachability, and "semantic" location. The delivery of both electronic and tangible goods will increasingly depend upon time-dependent addressing -- defined from the perspective of (mobile) computing devices and mobile users.

A provisional/draft survey document has been prepared to help developers become better aware of existing standards and activitied relating to interchange of name and address data. Topics covered include:

Principal references:

"Markup Languages for Names and Addresses."


SEARCH \| ABOUT \| INDEX \| NEWS \| CORE STANDARDS \| TECHNOLOGY REPORTS \| EVENTS \| LIBRARY