Internet survey. The first step was to carry out an internet-based survey to identify potential data providers and custodians. Special attention was given to on-line species and specimen data, with focus on biological collections and species and specimen data custodians. This information is evidently biased as it is restricted to data and information that is available on-line. But, as the objective is to develop an internet based information facility in order to quickly prove its utility and usefulness, it is important that existing initiatives be integrated while non Internet data providers organize and digitize their data.
Questionnaire. A questionnaire was prepared in consultation with GBIF. Its objective was to analyze a number of aspects in data providers and custodians such as:
Survey. The questionnaire was sent out to potential data providers and custodians known to GBIF and CRIA and to those identified in the Internet survey. A web site1 was developed with information about the project, inviting all interested players to answer the questionnaire that was made available on-line. Over 600 emails were sent to 254 institutions from 15 countries. More then 200 answers were received from 22 countries.
Personal meetings, phone conferences and hired consultant. Personal meetings were held at the American Museum of Natural History and at the New York Botanical Garden. Phone conferences were carried out with specialists from the Instituto de Investigación de Recursos Biológicos Alexander von Humboldt, Colombia, and the Instituto de Investigaciones de la Amazonía Peruana – IIAP, Peru. A consultant was hired to contact interested parties in Venezuela.
Studies. Specific studies were carried out on architectures, standards, and protocols used in different information systems.
Analysis. The questionnaires were received and analyzed and together with the IT studies were the basis for the production of the following reports:
Proposal for digitization of biological collections2
Proposed Architecture3
Data Sharing and Repatriation of Biodiversity Information: Setting-up a Collaboration Program with Collections from Non Amazonian Countries4
Dissemination of reports. All reports were made available on the Internet for comments and suggestions.
Technical meetings. A workshop and a briefing meeting were organized to present and discuss the reports and the results of the survey
A workshop was held with data providers and custodians of the Amazon region5 in Campinas in March, 2006. 18 participants from Bolivia, Colombia, Ecuador, Venezuela, Brazil, and Peru discussed with CRIA and GBIF first whether an information facility for the Amazon Basin is needed and second, if it is necessary and important, would those present be willing to collaborate. The answer was unanimous as to the necessity of structuring an information facility for the region and in the participants’ willingness to collaborate. A summary of the survey was presented and the architecture was discussed. Contributions expressed during the workshop and after (sent by email) were included in the feasibility study.
A meeting was held in Curitiba, Brazil, during COP 8 with 30 participants, mostly from non Amazonian countries. Here again the survey and architecture were presented and partnerships were discussed. The idea is to try and make digitization of Amazonian data a priority for collections held in Europe and North America. Here again, great interest was shown and worldwide collaboration seems to be feasible if the necessary resources are in place.
Feasibility study. Based on the surveys, the studies, and the contributions received at the workshop and technical meeting and through email, a feasibility study report was prepared and is being submitted to the Gordon and Betty Moore Foundation.
Architecture, standards, and protocols
When designing and implementing the network, it is important to consider the fact that ABBIF should:
promote free and open access to taxonomic information and non-sensitive specimen data;
enable that each data provider or custodian be fully responsible for his/her own data;
enable frequent updating managed by each data provider;
enable data validation;
assure that full attribution to data and information sources are given;
aim at strengthening local biological collections and data custodians;
be integrated to existing information systems at local, national, and regional levels; and,
be integrate to GBIF.
The proposed architecture for ABBIF is based on the study of existing systems that are disseminating species and specimen data through the Internet and on the answers received to the questionnaire by biological collections and data custodians as to what standards and protocols are being used.
The report on the architecture3 included the following elements:
Centralized, distributed and combined or mixed systems: main characteristics, advantages and disadvantages.
Standards and protocols: what is available, what is being used or has been adopted by networks of direct interest to ABBIF and what is the situation of local data providers and custodians concerning infrastructure and expertise.
Architectures: how data custodians and information systems of direct interest to ABBIF are being structured
