What if GBIF?

James Edwards makes a credible, convincing argument for digitization of museum data in “Research and Societal Benefits of the Global Biodiversity Information Facility” (BioScience 54: 485–486). I agree with most of his points, except the characterization of a recent essay by Raven, Wilson, and me (2004) as representing the view of digitization “doubters.” None of us doubts the essential role of digitizing data; each of us has advocated such work. Our point was that descriptive taxonomy continues to be neglected, even as digitized tools are developed. With many specimens falsely identified (Meier and Dikow 2004), most research funds allocated to phylogenetics divorced from classification, and predictions of imminent species extinctions, it is imperative that support be returned to revisionary taxonomy (Wheeler 2004). There are those—Edwards not among them—who do not recognize this crisis in taxonomy and who believe that supporting bioinformatics will address the “taxonomic impediment.” This is not so. Substantial (though insufficient) funds have been invested in bioinformatics for more than 20 years, yet the plight of taxonomy has worsened. Such investments, isolated from active taxonomic research, are little more than window dressing. Specimen data degrade unless species hypotheses are continually tested. That is most efficiently done through revisions, monographs, floras, and faunas. A suite of activities is crucial to the future of taxonomy: field inventories, revisions and monographs, and an array of data digitization. Digitization is urgently needed for museum data, specimen images, and descriptive literature. New digital tools for morphology and publication are essential. An international paradigm, following the vision of the National Science Foundation's Planetary Biodiversity Inventories projects, linked with digitization can revolutionize the field. It is not a matter of choosing between digitization projects and taxonomy; it is a matter of doing both. If the Global Biodiversity Information Facility (GBIF) succeeds, as it must, it will be a cornerstone resource. GBIF is needed desperately and now. If it did not exist by virtue of the vision and leadership of Edwards and others, it would have to be invented. I do not doubt the essentiality and worth of GBIF; I am an unwavering, enthusiastic supporter. I am no less committed to rebuilding the stature, infrastructure, and vibrancy of descriptive taxonomy. Trustworthy knowledge is the goal; descriptive taxonomy, the assurance of reliability, and digital tools, the means. A vision is emerging for a taxonomic cyberinfrastructure—a virtual instrument, observatory, and research environment for taxonomy (www.flmnh.ufl.edu/LINNE/)—driven by scientific questions, built on the strengths of museums and digital connectivity. The big questions of taxonomy are essential to the advancement of science and human welfare, and they are unanswered. GBIF and other digital initiatives have paved the way for this brave new vision of taxonomy, but they will ultimately succeed only to the extent that taxonomy itself does.