Centre for Biological Information Technology, The University of Queensland, Australia
Two major rate-limiting steps in developing interactive identification keys are the construction of character lists (constrained terminologies) with associated auxiliary images and glossary entries, and the harvesting of existing data. If these steps could be made more efficient, it would be possible to greatly increase the speed and ease of production of such keys.
The problem is one of standards and protocols. The TDWG-SDD standard has now provided a means for representing the vast body of existing but disparate data (existing matrix keys, existing pathway keys, existing phylogenetic data matrices etc) that could contribute to a developing key. But there are currently no standard character lists for most taxa, no standard protocols for sharing and exchanging such lists as could be developed, and no protocols that would allow high-throughput harvesting of key data.
The Key to Life seeks to address these problems. It comprises an XML-based store of keys (both matrix and pathway) and other data sets, and of nested, normative character lists for taxa. Key developers may download character sets and data matrices from the Key to Life in SDD format, and use these as the basis for constructing their own key. It is hoped that developers will also share keys by contributing keys to the Key to Life. The Key to Life, once populated, will also be useful for validating keys against existing data, and for on-the-fly delivery of identification keys to key users.