Portable ISO 19115-2 Open Source Developer's Toolkit (Bureau-wide Application)

Over the last few years, the ISO 19115 family of metadata standards has become the predominantly accepted worldwide standard for sharing information about the availability and usability of scientific datasets among researchers.  The U.S. interests in the ISO standard have also been growing as global-scale science demands participation with the broader international community; however, adoption has been slow because of the complexity and rigor of the ISO metadata standards. In addition, support for the standard in current implementations has been minimal. 

Principal Investigator : Stan Smith, Joshua Bradley
Cooperator/Partner : Chis Turner

In 2009, the Alaska Data Integration Working Group members (ADIwg) mobilized to jointly address common data integration efforts.  Beginning in 2012, ADIwg started to focus on difficulties associated with generating and exchanging metadata using the ISO family of standards.  Like a microcosm of the larger world, ADIwg partners vary in size from small nongovernmental organization research groups to state offices, universities, and large Federal bureaus.  This differential in size and technical capabilities among its membership brings with it the predictable diversity in metadata requirements and ability to provide the necessary technical support to its researchers. 

The ISO 19115 metadata standard is not a single standard but a family of ISO and Geography Markup Language Encoding standards.  Understanding this amalgamation of standards to a degree sufficient to generate a valid metadata record requires an investment of time well beyond what can reasonably be expected of any PI or end user.  These issues are not specific to ADIwg and its members; they are shared by many organizations striving to transition to the ISO metadata standard. 

After much discussion, the ADIwg membership decided to co-author an ISO software toolkit with an architecture that has the flexibility to achieve the following goals:

• Isolate the complexity, rigor, formatting, and terminology of the ISO standards from the PI and (or) data steward.  Users should not be required to be experts on the ISO standards to generate ISO metadata.
• Provide clear developer and user documentation.
• Design a layered architecture and provide developers with access to each layer. Developers should be able to enter the architecture at whatever layer best fits their needs. For example, they should be able to download the code library to do custom programming or post JSON metadata to a hosted JSON-to-ISO translator. 
• Design the architecture to accommodate incremental development cycles and add features without disrupting previous implementations. 
• Implement all code and services as open-source software projects using a GitHub repository (Alaska Data Integration Working Group, 2015a). 
• Use simple JSON for storing and transferring user metadata.
• Provide flexibility to write to multiple metadata standards, including ISO 19115-2:2009, 19115-1:2014, 19110, and possibly the Federal Geographic Data Committee’s (FGDC) Content Standard for Digital Geospatial Metadata. 
• Provide an online, interactive, client-side, end-user metadata editor written in JavaScript to gather and edit a user’s metadata and format it in the mdJSON format. 
• Host a publicly available version of the online metadata editor for end users to enter and edit metadata and then request a valid ISO standard metadata record to be returned.
• Write all code in popular, well-supported computer languages that are platform independent, royalty-free, and available through an open-source repository to encourage participation from and benefit to the widest possible user base.