Encoding Standards

Encoding Standards

After an analysis of key issues around encoding policies, (cite: Brown center) a decision was made that the encoding standard for this project should be Dublin Core. Initially, the guidelines of the Text Encoding Initiative (TEI, version 5) were considered a better fit because of its ability to handle a very broad range of formats and contents while, while enabling very rich information capture, particularly when dealing with lengthy resources requiring extensive transcription. In addition, its widespread adoption across a range of institutional users for a wide variety of purposes and fields, as well as its ability to be modified as information goals change over time, made it seem like an obvious choice.

However, closer examination to the specific particularities of this project with respect to factors such as resource features, audience needs, and intended user experience reveals that this might not be the case when looking at the overall picture. When we consider, for example, the physical properties of the final interface – relatively small screens on smartphone and other mobile computers – it becomes apparent that complex, difficult-to-read, or lengthy documents could not be properly read. In this light, clearer, more impactful and more compact materials are more likely to meet the realities of the technological environment. Similarly, when the particular needs of the target audiences are properly anticipated, encoding priorities shift once again. Since the primary goal of Digital Chinatown is to provide on-site, document-enriched experiences for non-professional audiences such as students, tourists, and community members, the degree of depth and complexity in information will be less than it would be for academic researchers and other history professionals. The envisioned user experience, mediated as it will be by mobile technologies, is still more different from the typical digital archive experience. Although the capacity to conduct searches will be included as a feature, the primary feature will be an ‘augmented reality’ experience, one involving immediate integration of images, documents and other resources in real time and space, and geospatial data serving to ‘trigger’ the activation of resources linked to the spatial position of the user. Furthermore, combining traditional search capacities with geospatial data, search results will be represented across a map in order to give the user an enriched experience with resources as they are situated in space and in time.

In the context of augmented reality experiences, it becomes more important that resources maintain fast and effective relationships among themselves and can be activated, on their own and in concert with other resources, with ease and reliability. At the same time, it leads directly to questions of practicality and the need to choose an encoding standard that will permit the construction and maintenance of such features efficiently and cost-effectively. Taken altogether, Dublin Core is clearly the best choice among encoding standards. All the metadata required to fulfill the above features can be encoded effectively, quickly and economically. As an international standard with continued and broad adoption across disciplines, Dublin Core is also an effective long-term solution for a stable and flexible encoding standard.

Developed and regulated under the auspices of the Dublin Core Metadata Initiative (DCMI), an international organization with the mandate to “provide simple standards to facilitate the finding, sharing and management of information.” A large part of its simplicity is due to the streamlined nature and compact nature of its metadata fields. With only fifteen fields, it lends itself to relatively straightforward training, standardization and implementation. For Digital Chinatown, all fifteen fields will be used in order to permit the maximum potential for search and integration features. Furthermore, when more than one category of metadata can be envisioned for some fields (i..e, Description, Relation, and Coverage), Dublin Core’s built-in capacity to permit repeating of fields will be used. Altogether, the following represents the Dublin Core fields that will be used:

1. Title

2. Creator

3. Subject

4. Description (General)
4. Description (Text)

5. Publisher

6. Contributor

7. Date

8. Type

9. Format

10. Identifier

11. Source

12. Language

13. Relation (Spatial)
13. Relation (Personal)

14. Coverage (Spatial)
14. Coverage (Temporal)

15. Rights

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License