Project Completion Report for:

Establishing Washington State's Geospatial Metadata & Clearinghouse To Support the National Spatial Data Infrastructure

Introduction

Appendix G: Metadata Creation Workshop Flyer, Agenda and Presentation Materials

Introduction

In August of 1995, Washington State Geographic Information Council (WAGIC) successfully submitted a proposal to the Federal Geographic Data Committee Cooperative Agreement program for a statewide geospatial metadata coordination project.

This report describes the creation and implementation of the infrastructure for Washington State's spatial metadata and clearinghouse. The report provides background information and discusses the issues discovered and their resolution during the course of the project.

The title of the proposal was "Establishing Washington State's Geospatial Metadata & Clearinghouse to Support the National Spatial Data Infrastructure".

Background

The Washington State Geographic Information Council is the statewide body responsible for the facilitation and coordination of the use of geospatial information in Washington State. Formed in 1982, the Council has fifty-two member organizations with representatives from federal, tribal, state, and local jurisdictions as well as the private sector.

The Council has historically focused on spatial data related issues. Activity to promote spatial information access and data sharing was confirmed as a high priority goal during the Council's 1995 Strategic Planning Session. During subsequent discussion it was agreed that a successful grant proposal could provide the appropriate focus for realizing these goals.

WAGIC facilitated a partnership between: Washington State Department of Information Services, Department of Natural Resource, Snohomish County and Central Washington University to write the grant proposal and implement it. Near the end of the project the University of Washington's Geography Department and Libraries Map Collection Department also played important roles.

The focus of the project was to cooperatively develop and manage infrastructure for geospatial data access according to a state-wide geospatial data initiative and the National Spatial Data Infrastructure.

The ultimate goal of the project was to improve access to spatial information about Washington State through cost-effective utilization of the many spatial 'datasets' that currently exist. It has been estimated that sixty-five to seventy percent of the operational costs associated with a successful Geographical Information System (GIS) is related to obtaining the spatial data for the system.

Our State is typical of most and has thousands of spatial datasets that have been created over time for a variety of purposes. The focus of the project was to provide a standard way to describe the data that is available today and then make those descriptions (metadata) publicly available through the Internet. Successfully implemented this readily available reference will help reduce the money spent developing data that already exists and allow each jurisdiction to focus it's data creation resources on it's unique needs.

• Create a publicly available, useful infrastructure for collecting metadata descriptions and disseminating them over the internet
• Metadata descriptions available through the NSDI Clearinghouse Gateway
• Several organizations utilizing FGDC Metadata Content Standard (MCS) to implement their metadata
• An understanding of how to utilize MCS and Clearinghouse tools to put information on internet

• Data descriptions of: five statewide data themes, twenty plus county data themes
• A Metadata collection tool
• Create a Clearinghouse Node
• Mapping FGDC MCS to proposed state architecture (see Note 2)

For details of the project proposal please refer to "Establishing Washington State's Geospatial Metadata & Clearinghouse to Support the National Spatial Data Infrastructure". A copy of which is available at the WAGIC Website: www.wa.gov/gic/

Implementation Approach

Each of the partners played key roles in the project success. Department of Information Services was responsible for the development of the metadata collection tool. Department of Natural Resources and Snohomish County use several different tools for creating metadata descriptions to give the project team experience with several approaches to collecting metadata. Central Washington University did an initial investigation of clearinghouse software (WAIS and Isite). University of Washington provided technical support for creation of the Isite Clearinghouse node.

Benefits

By embracing the NSDI Clearinghouse concept and underlying metadata standards the State now has an information access resource that will prove useful to the public and GIS professionals. It is important to note here that this is just a start. This project has laid down the foundation for a very rich information resource. By creating the infrastructure and seeding it with examples of both statewide and local coverage’s this project has paved the way for cost effective use of existing spatial information resources.

The importance of standards is fundamental to a sophisticated approach to information access. The value of implementing a national standard is that we 'instantly' become part of an national registry of connected spatial information resources. This soundly positions Washington with not just a standards based approach, but more importantly, on standards that are being embraced by neighboring states and states through out the country.

Through the use of the Clearinghouse gateway users can determine what information is available about Washington State irrespective of where that information resides. An information query submitted through the Clearinghouse gateway is sent to all the registered Clearinghouse nodes (including the Washington State node). It is important to note that spatial information about Washington State is created and maintained in numerous places outside the state.

Effective access to and use of this information is made possible through this global network of Clearinghouse servers and the query capabilities of the Clearinghouse gateway. The Metadata Content Standard makes this all possible. By identifying a standard set of elements and names for those elements it makes it possible to store the information in a way that enables the powerful Clearinghouse query capabilities.

This project provided the opportunity for four WAGIC member organizations to develop an in-depth understanding of NSDI Metadata Content Standard (MCS) and related initiatives (Clearinghouse, Framework). The value of this is that there is now a core of local expertise regarding how you collect, maintain and disseminate spatial information utilizing the NSDI Clearinghouse approach. This combined with the fact that the supporting infrastructure is now in plane positions our State to grow a truly useful spatial information resource.

Spatial data users have realized the following from the successful implementation of the project:

• There is an increased awareness of the importance of metadata, particularly the value of metadata done to a common standard
• They now have on-line access to spatial data descriptions
• They now have some 'local' resources to help them understand how to create MCS compliant data descriptions and make them available on the Internet
• The infrastructure is in now in place to support the Clearinghouse. This provides a home for, and immediate, widespread usefulness, to an organization's metadata

A metadata Collection Tool

This metadata collection tool is a Microsoft Access application with the following major features:

• Supports the creation of MCS compliant metadata descriptions
• Rich help environment (including context sensitive access to FGDC MCS Workbook 'Green Book' examples and definitions).
• Navigation via FGDC graphical interface
• Flexible environment that encourages use of standard and continued upgrade of description content

Development of this tool took nine months for the beta version. Testing and refinement was done over an additional four month period. The basis for the tool is an environment that was created to provide an automated 'cross-walk' between the spatial portion of a (then) proposed Washington State Common Data Architecture (see note 2) and the MCS.

GMMS is fundamentally a flat file implementation of the MCS but does provide multiple tables in the database to accommodate multiple iteration of certain descriptive attributes within the same metadata description and/or across several metadata descriptions. The tools supports the use of template descriptions to facilitate the creation of a set of metadata description that have a large number of common attribute values and differ only on some key elements. A template with the common attributes is created and copied each time subsequent new descriptions are created. Many organizations find that eighty percent of the metadata falls in this category. Consequently this feature is very useful.

Snohomish County utilized the GMMS tool in creating their metadata for the project. Their experience with the tool suggests that it would have benefit for organizations that were looking for a standalone database environment.

GMMS is available without support in two versions. Source files can be provided for those with a licensed copy of MS Access 2.0 or in a 'run-time' version for those without MS Access. Platform requirements are Windows 95. (Warning: MS Access 2.0 software platform is the only one supported by GMMS - use of GMMS in combination with any other version of MS Access could result in alteration of the installed Access.ini file(s) and lead to unpredictable problems. (See Appendix A for copies of GMMS and selected screen prints)

In our current Metadata Creation Workshops one of the collection tools we are featuring is a vendor supported product (Spatial Metadata Management System from Enabling Technologies, Richland WA) that got it's geneses from early GMMS architecture. The value of this approach is that potential users are able to purchase varying degrees of support and can obtain appropriate version support.

The real value of this activity was the eventual understanding that it was not appropriate for WAGIC to provide support for the multiplicity of MS Access environments.

CNS and MP are the editors provided by the FGDC to validate that metadata descriptions fit the tenents of the standard. These are DOS command line utilities and can require lengthy obscure command line inputs each time they are invoked.

To increase our metadata processing efficiencies we created a Windows interface to the CNS and MP. The Windows interface noticably reduced errors and made the process easier for non-technical types to manage. A copy of the Windows Interface to CNS & MP is included in the appendix (See Appendix B for copy of Interface and selected screen prints )

Features of the Interface are as follows:

• Allows user to invoke the DOS processes from within the Windows  environment
• Prompts for target and destination file names
• Automatically brings up error file at completion of process to facilitate error reduction
• Helps manage file naming conventions

This application was developed in Visual Basic 5.0 for 486 or newer PC. The application will be made available for ftp down load at the WAGIC Website.

Twenty-eight metadata descriptions were created that are compliant with the MCS. Five of these are statewide data coverage's provided by Department of Natural Resources and the balance are county based coverage's provided by Snohomish County.

The tools used to create the metadata documents varied. DNR used an in-house word processing template (Word Perfect), Snohomish County used the GMMS tool. Regardless of tool used, each description was saved as a formatted ASCII file and subsequently run through CNS and MP to generate the files necessary for posting to the Clearinghouse Node. Typically, there were several editing iterations necessary to get the desired output from MP. The number of iterations tended to drop once an understanding of what caused them was identified. MP is 'fanatic' in the way it enforces indentation and it is not obvious how it will deal with recurring elements within the MCS.

An important element in our approach to starting organizations on creating metadata is the Basic and Working subsets of the MCS that we use when introducing potential metadata contributors to the WAGIC metadata and clearinghouse initiative.

Early on in the project it was apparent that we needed a way to introduce potential contributors to the MCS. Because of resource constraints many of our potential customers simply would not embrace the standard if it meant full implementation out of the starting gate. A Basic and Working subset of the MCS were identified and presented as a way to start using the MCS (and subsequently the Clearinghouse). Encouragement was given for each organization to make use of the MCS, as resources would allow.

The main focus was to get them started with the idea of building metadata within the structure of the MCS so we can make it available via the Clearinghouse. Each contributor was and will be encouraged to adopt MCS at a level of detail that makes business sense to them. (See Appendix C for selected example)

This keystone activity of the project was comprised of three significant tasks: collecting and creating MCS compliant metadata descriptions creating the platform and creating a clearinghouse operating agreement and partnership. Creating the metadata is discussed above. Creating the platform and environment was a relatively straightforward technical task. Creating the operating partnership and agreement while not difficult was time consuming.

The importance of this process was that we had the opportunity to explore the issues associated with the setting up the software and with creating a workable agreement under which Clearinghouse can grow and prosper as a spatial information resource for professionals and the general public.

Over the course of this project the preferred platform for Clearinghouse nodes changed. Original focus was on providing a WAIS type clearinghouse node but it became apparent that adoption of the more powerful Isite software was appropriate. Central Washington University conducted initial exploration of the WAIS and Isite.

The installation and configuration of the software environment was fairly straightforward thanks to the work done by the FGDC (Doug Nebert, ET El) in packaging and providing training tutorials and workshops.

The actual implementation of our initial prototype clearinghouse was done in partnership with University of Washington Geography Department. There, we were able to utilize a server that had Isite software installed for a 'How to Set up a Clearinghouse Workshop' taught by Doug Nebert. (See Appendix D for WAGIC Clearinghouse screen prints)

This temporary arrangement allow us to test our proposed plan for administering the Clearinghouse and made us familiar with software configuration and testing processes. The experience positioned us to create a meaningful Clearinghouse operating document and partnership with University of Washington Libraries Map Collections.

The original parties of the operational agreement include WAGIC, WADIS and UW Libraries, Map Collections Department.

A formal agreement was developed between Washington State Department of Information Services and the University of Washington Libraries Map Collections Department to permanently host the clearinghouse platform at the UW.

The Clearinghouse operating document outlines the following:

• Purpose and business reasons for the Clearinghouse
• Defines broad operational goals for the Clearinghouse
• Sets up a WAGIC Clearinghouse Steering Committee with operational policy making responsibility
• Defines six roles that need to be filled for successful long-term operation

For details of the proposal/agreement please see Appendix E.

Early on in the project planning it was apparent that a comprehensive communication and education strategy would be necessary for project success.

Our coordinated communication campaign had two major components; promote the NSDI initiatives and education that focused on the details of MCS and Clearinghouse. Promoting NSDI initiatives provided a forum and level of detail useful to managers who wanted to understand why they should become involved. The detailed education focus on the MCS and Clearinghouse was useful for those who needed to understand the 'how' part of implementing our approach. We applied this two component approach through a variety of public forums.

The Council focused its promotional materials on Spatial Data Sharing and the value of the NSDI initiatives. The project and related activities we heavily featured in: the Councils’ three-fold brochure "Washington State' Geospatial Solution"; an innovative, award-winning video titled "The Newest Frontier"; and the professionally designed traveling exhibit featuring Council activities.

The project and associated NSDI initatives were featured in the following forums:

• Periodic project status reports at WAGIC meetings
• As part of WAGIC Exhibit at following conferences:

• Washington State GIC '96 Conference - November '96
• NWArc Info User Group Conference '96 - October '96
• Information Processing Managers Association Fall Forum - Sept '96
• County Commissioners Annual Conference - June '96

• One on one contact with people that are actually responsible for creating the metadata - these were very important contacts and typically resulted in the programs strongest supporters
• Utilizing Internet early on we created a Clearinghouse/Metadata page on the WAGIC Website to show example(s) of what we were shooting for. This allow us to point to Washington State examples of metadata descriptions and give people a sense of what was possible by visiting the NSDI Clearinghouse Gateway and demonistrating its value.

The Council co-sponsored several FGDC MCS and Clearinghouse Workshops

• Video-teleconferenced session to six sites around state
• MCS and Clearinghouse sessions at GIC '96 Conference (see Appendix F for Conference program)

The Council created and delivered a "Metadata Creation Workshop". The purpose of the class was to serve as prototype for regional delivery around state to encourage and build information content of Clearinghouse. This one day "free" class provides an overview of the various types of metadata collection tools and hands on experience with the tool of the attendee choice. Attendees actually post at least one example of their metadata to the clearinghouse by end of the class. While the class is offered 'free' of charge, the 'Price of Admission' is to bring enough information to the class to complete at least one metadata description to post on clearinghouse. (See Appendix G for class flyer, syllabus & presentation materials)

As is often the case the value of the project lies not only in the 'deliverables' but also in the lessons learned.

Do not underestimate the process and organizational side of creating a Clearinghouse node. Long term successful operation of a Clearinghouse node as a useful information sharing resource is more than a technical exercise. (See "Successful Clearinghouse section of Clearinghouse document, Appendix E) Identifying the appropriate organizational structure and creating a partnership to support it are fundamental to the long-term success of the clearinghouse.

Be prepared to support a subset of MCS (while encouraging more robust descriptions). This will provide a pathway to get organizations over the metadata start-up hurdle. Look at the FGDC minimum searchable set or see WAGIC's Basic and Working subset as example of this approach. There is a need to provide organizations with a flexible approach that allows them to move from where they are today (often with no or limited metadata) to full FGDC compliance (where appropriate) through a flexible migration path.

It is not enough to use a ‘build it and they will come’ approach. You need to be prepared to educate and promote the effort at both the managerial and technical levels for potential contributors. The concept of metadata can be very tedious and the benefits of a clearinghouse are not immediately clear. Be prepared to employ a variety of promotional techniques and to sponsor/create metadata classes.

The MCS is not presented by the FGDC as an architecture for implementation rather it is a document that identifies what should be included in a metadata description. While the MCS describes 'what should be included', potential tool developers need to plan on investing many hours in determining how to develop an application to support it. For instance, it is far too easy to mistake the numbering convention as a hierarchical index for a database. At least fifty percent of the development time was taken with understanding the MCS requirements and sometimes troubling inconsistencies.

In the end the significant issue with GMMS became the inability to create and support different versions of the applications to keep up with the release of multiple version of MS Access. At last count there were six different versions of MS Access (2.0, 3.0, 4.0, 7.0, Office '97, WFW versions) with limited upward compatibility. Consequently, we elected to provide only unsupported copies of the application.

The Council and its partners will continue to implement a spatial information infrastructure in Washington State by focusing on two aspects of the NSDI: Enhancement of the information content of the Clearinghouse and the Creation and coordination of Framework Data initiates.

There are currently plans to sponsor four regional metadata workshops this spring and summer that will further understanding of FGDC MCS and the tools available to create compliant metadata. The focus is now on increasing the information content of the Clearinghouse (more data descriptions from more and more organization) until the Washington State Spatial Information Clearinghouse becomes ‘the’ place to go for the states spatial information.

There are several Framework Data creation projects underway in Washington State. The Council’s Framework Management Group serves to provide a forum to coordinate the projects. The group is actively pursuing ways to incorporate and leverage the Clearinghouse with these projects.

Coordination will continue with various emerging Clearinghouse efforts within the state. A metadata effort centered on the Olympic Peninsula region and headed up by the University of Washington through the Olympic Natural Resources Center will be providing metadata for the Clearinghouse. Metadata representing the results from a recently completed statewide GAP Analysis is being prepared and will be posted to the Clearinghouse with live links to the underlying data for download over the internet.

Creating a Spatial Information Access Track for upcoming statewide GIS Conference. Focus will be on Framework Projects and their value and relationship to the Clearinghouse. Continued promotion and education efforts will focus on how to effectively use the Clearinghouse both for data contributors and users.

These activities will continue to be high priority items on WAGIC’s agenda. There is recognition of the need to move on to the next phase of our implementation of the NSDI initiatives. There is now a heighten sense of the value of our approach. Both for making GIS a more effective decision support and management tool and for improving public access to Washington States spatial information.

Note 1

FGDC/NSDI Initiates: The purpose of the FGDC National Spatial Data Infrastructure (NSDI) Competitive Cooperative Agreements Program is to facilitate and foster partnerships and alliances within and among various public and private entities to assist in building the NSDI.

The NSDI consists of policies, standards, agreements, and partnerships among a variety of sectors and disciplines that will promote more cost-effective production, ready availability, and greater use of high quality geospatial data.

At the time the project proposal was put together, Washington State Department of Information Services was investigating a proposal to adopt a statewide IT data architecture. The proposed Common Data Architecture (CDA) was a thoughtful approach to creating a large-scale data architecture for the State. The broad scope of the proposal included a number of data models specific to various data themes including a Geospatial data theme.

When the project proposal was written there was significant interest in finding a way to map this candidate architecture to evolving Federal level spatial data standards. Specifically, develop an automated cross-reference from the spatial component of the proposed Common Data Architecture and the Federal Geographic Data Committee's Metadata Content Standard.

In the end, the proposed Common Data Architecture for Washington State was not adopted but this project benefit greatly from the conceptual work done in creating the Common Data Architecture initiative.