I found that the most difficult aspect of this assignment was finding an appropriate scope. I began with an ambitious proposal to adapt the New York Heritage metadata schema into an application profile that would be capable of transitioning into a linked data environment. However, I soon realized that my undertaking was a several-month project that did not fit within the scope of this assignment. Back to the drawing board.
Once settling an the domain of golf discs, I came to appreciate the balance between being comprehensive and realistic about the implementation of an application profile. Include too many elements and users will be overwhelmed by the amount of information; include too few elements and it becomes difficult to gain a full picture of the domain. I channeled the simplicity of Dublin Core in developing an element set to describe golf discs, asking myself: what are the fundamental components of a disc that would be helpful to manufacturers, retailers, and players?
Many of the difficulties in maintaining this simplicity stem from the messiness inherent in any domain. Even if there are a critical mass of typical items, the minority outliers contribute disproportionately to this messiness. In the case of this domain, there are a few golf disc manufacturers who use an entirely different set of identifiers to describe the flight characteristics of their discs. While I have attempted to incorporate this diversity through the inclusion of the discStability field, there may still be discs that do not fit neatly into this schema. Not to mention the fact that new discs are being developed at a rapid rate. My application profile would have to be updated over time to address these outliers and new discs. Maintin
In the end, as metadata developers, we must do our best to describe and make order from the messiness of everyday life.