Automating metadata for dynamic datasets

Data and metadata are intrinsically connected. Both are vital components of a complete dataset and neither should be viewed as separate entities. Real-time and hourly data collections are dynamic, driving the need for amended metadata to record these types of observations. Metadata is simply data about data that documents the characteristics of the data, such as condition, quality, and content. There is a vast array of various metadata standards such as the ISO 1911* series for Geographic information, the North American Profile of ISO (NAP), the Federal Geographic Data Committee (FGDC) standards, Directory Interchange Format (DIF) metadata, and library metadata standards such as Dublin Core (DC). Each of these standards is also available in a variety of formats, such as text, HTML, XML, and FAQ views. Creating and maintaining various types of metadata for dynamic data can prove to be a daunting task. Automating metadata creation and maintenance processes is an approach to combat the possible resource drain that can be caused by constantly evolving metadata, especially if that metadata is required to be output in a variety of formats and standards. Metadata can be automated by collaboratively using a variety of input sources such as multiple databases, reports, and active sensors to create and maintain dynamic metadata in a variety of formats and standards through the use of extensible Stylesheet Transforms (XSLTs) and extensible Stylesheet (XSLs). Directly creating various metadata formats from the input sources, rather than translating between various standards, aids in preserving information. In legacy data cases, XSLTs can be used to translate between metadata standards. The transformation process could be triggered by updating or altering the input data sources. Various trans-forms and/or stylesheets would be selected, depending on the trigger, applied to the altered input, and result in the generation of multiple records in a variety of standards and formats.