Preprints of the Metadiversity Conference Proceedings

  Session 7: Working Groups

Reports from the Working Groups

RICHARD T. KASER, NFAIS Executive Director,
Metadiversity Principal Investigator

At the end of the second day of the Metadiversity program, participants were formed into groups of four, and then these groups were combined into groups of eight, to list issues for further discussion. The recommendations from these small working groups were compiled and merged into five discussion topics.

1. Leadership & Consensus Building, including national vs. global issues, organizational framework (infrastructural issues), interconnection of efforts (across domains, disciplines, and organization types), and how to achieve support from individual organizations (including ones own organization) and related communities.
    
2. Technology, including how to establish priorities, how to assure distributed but interoperable systems, how to handle data archiving and assure long-term data access, and data security.
    
3. Standards for Biodiversity Data, including taxonomy (controlled vocabulary), thesauri, models, and tools. This group also touched on the larger issues of how to involve professional communities in developing interoperable systems.
    
4. Funding & Economics, including funding strategies, developing incentives and rewards, business models, test beds, and business plans.
    
5. Users, including identification of user groups, their requirements, standards, privacy issues, etc.

Participants in the symposium then registered to participate the following day in one of these discussions. There were four discussion groups in all. A single group considered both topics 4 and 5.

Facilitated Discussions

The four discussion groups were facilitated by: Anne Frondorf, Program Manager, National Biological Information Infrastructure; Bonnie Carroll, President, Information International Associates; Gail Hodge, Information International Associates; and Susan Warner, President, The LEAD Alliance, a consulting firm.

Each group reported its recommendations in a concluding plenary session of the symposium. Results are summarized below.

Recommendations

At the highest level, all of the reports and recommendations can be said to have focused on infrastructural concerns. The symposium program as a whole had stressed the range of agencies, data types, and systems already in existence. Many presentations had also emphasized the need for developing interoperable systems so that users of biodiversity information could have the capability of seamlessly crossing organizational and subject lines in order to obtain the information they need.

But what is the motivation for agencies to work together to create an interoperable system that exceeds the mission of their individual organization? How to inspire cooperation? And how to develop and win acceptance for standards that would achieve these ends?

The Call for Community

All of the Metadiversity working groups picked up on this theme. The Leadership and Consensus Building Group, for example, focused on how to get administrative officials (both in the government and at individual agencies and non-governmental organizations) to buy off on the need for cooperation. Another discussion group originally set up to discuss technological matters redefined its charter to discuss the culture of the scientific community and how to get working scientists to buy off on the concept that capturing data about data sets (metadata) is even important.

The Standards group considered the issue of how to get organizations to develop and accept metadata standards for their individual projects. And finally, the Economics/Users group ended up proposing activities that would result in better coordination of efforts throughout the world. The overall outcome of these group discussions, therefore, appears to be that consensus building within the entire community of biodiversity-related agencies itself needs to be a high priority.

The Metadiversity symposium aimed to produce a call to action. By the end of the conference it was clear that the call that emerged from the discussions was a call for community. Before we can proceed to solve the grand challenge of biodiversity information management, those who must be a part of the solution must first recognize that they are part of the "biodiversity information community." Then they must accept that they need to work with each other to create a common solution. But how will this be achieved?

Consensus-Building at High Levels

The "Leadership" group focused in particular on the need to develop consensus across disciplines, across international boundaries, and across organizations.

Emphasizing that the system needs to built with consensus from the bottom up and not by imposition from the top down, the group recommended a four-part strategy:

• Define biodiversity communities
   
• Define the scope of the metadata framework
   
• Proactively communicate across communities
   
• Build cross-community linkages

To implement this strategy, the group felt it was critical to make a convincing case to leaders of the value of metadata. It suggested that one way of doing this was to develop a prototype system that would show the benefits and demonstrate the capabilities of a fully deployed biodiversity information management system. Specifically, the group proposed that CIESIN, with international endorsement (e.g., from IUBS), should be asked to select one or several queries/examples that would include a range of organisms and geographical coverage. The BIOECO working group of CENR would advise on the selection of examples.

Development of such a prototype would also depend on the prior definition of an information model, which would include all the various kinds of data necessary to support the full-fledged system–in other words, a scope description of the metadata framework for biodiversity information management.

The group proposed that NBII should be responsible for contracting with a third party to both define the framework and to obtain the sample content to demonstrate the viability and usefulness of the desired system. In order to secure grassroots support, NBII would distribute a draft of this plan for community input.

Proactive communication across communities was also considered imperative. Specific recommendations were:

• Approach relevant umbrella groups to carry message to their memberships–for example, AIBS for professional societies, CENR for governmental agencies.
   
• Encourage NSF to include metadata as part of the products it funds.
   
• Encourage the development of a biodiversity informative consortium by communicating through professional societies (e.g., AIBS).
   
• Encourage NBII and CANBII to establish a resource site, including tools and resources.

As part of this effort, it was also considered important to develop creative relationships with non-traditional partners–i.e., non-governmental organizations, including private industry, not only as funding sources but also as contributors to the development of system plans.

And finally, the group suggested that NFAIS sponsor a follow-up meeting to Metadiversity to continue the dialog that began here.

Consensus at the Grassroots Level

While the first group looked at building consensus at high levels, a second discussion group approached the subject by focusing on the need to have a buy-in from practitioners (scientists, researchers, and scholars) who produce and subsequently use (or could use) data related to biodiversity. This group first observed that the development of an integrated biodiversity information management system was more limited by cultural factors than by technological factors. Researchers currently have few incentives and little motivation to provide documentation for the data sets that they create. Such documentation, however, was considered necessary to support and encourage information reuse.

Cultural factors that were identified by the group included the academic reward system (which currently does not require the sharing of data), funding mechanisms (which do not currently require that data be submitted with the results), instrumentation functionalities (which do not automatically record data about the data being computed), and professional standards (which do not universally extend to data documentation).

As recommendations, this discussion group called for a combination of educational efforts aimed at the research community and for the development of incentives that would encourage researchers to create and capture data about their data sets. The group also suggested that colleges and universities should extend the current system of rewarding researchers for publication to also credit researchers who publish data sets. The group also recommended that funding agencies require grant recipients to submit data as part of the results, and that the manufacturers of laboratory instruments be persuaded to incorporate software that requires the capture of metadata about the tests being run.

To encourage the above developments, the group proposed the creation of an Information Technology Resource Center. The center, which could be sponsored by the NBII, a professional society, or even a journal publisher, would develop and promulgate technical requirements for the collection and reporting of data, define specific data that should be collected, and provide resources, demos, etc.

This group also discussed the need for a browser/search engine that "understands biology," i.e., a search interface that can handle the technical vocabulary and can be used to qualify resources. It specifically suggested that one R&D priority should be the development of the basic ontology to support a biodiversity knowledge base.

This group echoed the first discussion group by concluding that in order to achieve these goals it would be necessary (and desirable) to bring the various communities together who need to support such developments.

Involving Users

While the second discussion group was deliberating about cultural factors affecting the user community, a third group was specifically focused on discussing user needs. Noting that the user community is diverse–consisting of multiple user types and "market segments" that have differing but overlapping needs and metadata requirements and includes K-12 students as well as professional research scientists–this third group stressed that proactive means of assessing user needs and collecting feedback are very important.

In order to involve users, the group recommended the following steps be taken:

1. Create an international association of biodiversity data providers and user representatives.
    
2. Address the issues of economics/politics, regarding open international access to data/Metadata.
    
   Academic/scientific communities must develop a consensus position–perhaps facilitated by the National Academy of Science or National Research Council.
    
   This biodiversity data community should propose an holistic assessment of user needs, including both current and potential users.
    
3. Metadata must be extensible for future use based on need. It must be an evolutionary process. It must be proactive.

In addition, the group proposed that funding sponsors–as well as core members of such a group–be identified.

The group also proposed that another workshop or symposium be held to:

• identify gaps in information needed to support users
   
• identify places user needs are being met and how they are being met
   
• bring in domain expertise in user assessment (library and social science)
   
• identify populations of priority potential users
   
• include Information Industry representatives to assist with value-added assessment

Economic Factors and Incentives

Given the size and scope of what could conceivably be defined as the "biodiversity community" (which in its ultimate sense could well include most government agencies and a whole range of industries), economic factors were perceived as an important area for discussion. Because the private sector needs to play a role, it cannot be assumed that all information that should be included in such a system can be provided for free. Thus the challenge in this area extends to considering intellectual property issues. A fourth discussion group took up this topic.

This group defined the problems as:

• Funding mechanisms differ for industry, academia, and government. There are three different business models involved: for-profit, cost recovery, and public domain.
   
• There is no funding for inter- and intra-sector collaboration.
   
• There is no funding for international liaison for metadata (e.g., IABIN).
   
• Often there is no direct funding for data management and metadata in government organizations or as a requirement of government grants.
   
• Bio-eco metadata communities cross many sectors, sometimes working at cross-purposes and competing for funding.
   
• Intellectual property rights issues affect funding approaches. This may create an increasing divide between information "haves" and "have-nots."
   
• There is little understanding of the user community and how users value information.
   
• What needs to happen?

The group envisioned the following ideal situation:

There exists a sustainable, financially stable, transnational, and cross-sector group to coordinate the creation, support, and promotion of bio-eco metadata (such as the Global Biodiversity Information Facility–GBIF).

There is resolution on a global scale of the intellectual property rights issue with respect to bio-eco metadata as well as the existence of a continuing mechanism for managing requirements for multiple property rights.

Bio-eco metadata is universally available in holistic environments that integrate elements of business models of for-profit, cost recovery, and public domain sectors.

Bio-eco metadata from within and across sectors is universally available in a coherent fashion.

There are funded mandates to government agencies for metadata creation and management at the international, federal, state, local, and tribal levels

How to get there? The group recommended these "bold steps":

1. International, cross-sector body: Establish ongoing national funding for bio-eco metadata activities and interests within an international, cross-sector body such as GBIF.
    
2. National dialog on intellectual property rights: Fund a national dialog encompassing diverse views on international intellectual property rights as they relate to bio-eco metadata (e.g., through NFAIS facilitation).
    
3. Development of an economic model: Fund R&D study to develop an economic model for making bio-eco metadata universally available in holistic environments that integrate elements of business models of for-profit, cost-recovery, and public domain sectors (e.g., a Harvard Business School study funded by the Melon Foundation).
    
4. Support for and promotion of existing successful activities: Fund the enhancement and promotion of existing successful models of cross-sector collaboration for bio-eco metadata.
    
5. Support for government agencies and government-funded activities: Put in place funding and incentives for government agencies (at all levels) and for recipients of government funding to create, manage, and share bio-eco metadata.

All of this, again, implies the need to achieve a consensus among the leaders of many organizations. However, this same discussion group also stressed the need to recognize that in the end it is all about users and their need for information. Thus this group reaffirmed that we also need consensus building from the bottom up, including the close involvement of the user community in defining what a distributed biodiversity information management system needs to do.

Standards Development

Finally, the fifth discussion group considered the need for standards as a unifying element in the creation of an interoperable biodiversity information management system.

Their recommendations were:

1. Continue prototypes and test beds for interoperability based on existing and emerging metadata standards
    
2. BRD, NCEAS (and any other relevant stakeholder organizations within the BioEco community) should convene a working group to develop a report/document and "tool kit" that will help our community to develop metadata standards.

• The document and tool kit would inform community members about resources available to help in establishing development and maintenance processes for metadata standards.
   
• The document should recognize that the biodiversity community is an assembly of smaller communities, and that these sub-communities have different information characteristics (e.g., diversity of data concepts and data distribution/partitioning) and culture. This means that particular standards development mechanisms might not be appropriate for all communities. The recommended best practices should reflect this heterogeneity.
   
• The document should include a time line describing milestones in standards development.
   
• The document should include recommendations about identifying, informing, and engaging stakeholders, such that the standards development process is open and representative of the community.
   
• The document should emphasize that every standards development effort must have an explicit scope, goal, and audience.
   
• The tool kit should make standards developers aware of the resources, particularly technical and human, that are available to facilitate the processes of standards development and adoption.

3. EPA (Bargmeyer), CAS (Blum), and TNC (Howie) will collaborate to enter a manageable number (~10) data elements into a data registry modeled on that of the EPA Environmental Data Registry.
    
4. Create a locator/registry of biodiversity organizations and/or data providers, perhaps similar to the Taxonomic Resources and Expertise Directory (TRED). This will help the community members become more aware of each other and will also help in identifying stakeholders in standards-development activities.
    
5. Continue work on shared vocabularies, including: create a registry of relevant vocabularies and thesauri, and do research into and develop prototypes of interoperability mechanisms for thesauri.
    
6. Tell BRD and NSF that an NCEAS-like center should be established to facilitate metadata and data synthesis. This center could then be used to host standards development and maintenance activities. At present, member organizations do not have the resources (staff time) to give standards development and maintenance the attention they require.

Summary of the Recommendations

In general what we saw being proposed was the creation of a centralized coordinating body, appropriately funded, to first identify and define the community and then conduct outreach, community-building, and support activities.

Previous

Copyright © 2003 NFAIS. All rights reserved. No part of this product or service may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written consent.

Privacy Policy