DRAFT
Global Nomenclators Architecture (GNOMA)
Date: August 28 2008
Location: Sale the Baleine, Pavillion de la Baleine at the Jardin des Plantes at the Muséum national d'Histoire naturelle, Paris, France
Chairpersons: Paul Kirk
Agenda
0900-1230 - Introduction and discussion
Introduction and Rationale: Why do we need a coordinated and global nomenclatural infrastructure?
Perspectives from within the nomenclators. Paul Kirk and Jerry Cooper
External perspectives from other initiatives. David Remsen, Yde de Jong, others
Discussion:
What do the nomenclators want to do for GBIF, EoL ... anyone else ... in terms of automated services?
What standard methods and data elements should be supported?
What are the use cases that should be supported?
How much should be code independent/dependent
Coordination and Visibility
Common index of Nomenclators (GNI)
What subset of metadata should be supported?
What reconciliation and de-aliasing should this index support
How should it be governed.
Demo of current GNI
1230-1330 - Lunch (Working lunch?)
1330-1630 - Discussions
How
Techical
Social
Financial
Short term vs longer term
1630-1730 - Summary and Wrap up
Recommendations
Outputs
1730 Adjourn
Topics for discussion in a Nomenclator-oriented meeting.
What do the nomenclators want to do for GBIF, EoL ... anyone else ... in terms of automated services? We need to decide on a standard list of Web Services that the nomenclators should provide. At the moment the IF ones are a bit clunky (we were the first so not unexpected) and sort of mirror the human web site – as such there tends to be a bigger packet of information than is perhaps ideal. Big challenge is to try to get a Code independent set of Web Service Definitions and probably some ‘BioCode’ terminology with both Botanical Code, Zoological Code and whatever else Code equivalents.
Should GBIF commission a cross-code task group to develop this code-independent set of definitions from which a set of implementation practices can arise? Who would be appropriate members (aside from the recipients of this document? ). This document would address the first question of what the nomenclators wish to do for potential users via their services.
Outcome would be a set of consensus web service methods
Do any of the TDWG TaxonAPI fit the possible needed service methods?
2. We need a Global Nomenclator Architecture (“GNOMA” - J.A.Cooper) – “an agreed usage of TCS to expose nomenclator data via TAPIR and via LSID mediated resolution of the appropriate TCS vocabulary content”. It then needs GBIF (or some other resource’) to provide a Name Resolution Service – for example so potential LSIDs can be returned from just a bunch of names. It could also manage any overlap between the nomenclators (e.g. because there is strong evidence that the microsporidia are fungi their names are now included in Index Fungorum).
Questions for discussion
What data should nomenclators expose?
First, to a Name Resolution Service and subsequently via the web services decided in 1.
AGI is done (almost) for Fungi but how far have the other groups progressed?
What is current interest among nomenclators for the All Genera Index? It seems there is consensus in the value both as an common good in itself and as a first step at a common indexing exercise.
We (IF Partnership) agreed that for any web site sat on top of the AGI with a public facing interface that only GenusName+Authority+YearOfPublication+Family (or next higher ‘real’ taxon)+Kingdom should be displayed with clickable links to the IF web site; for internal use the full dataset could be available. IF services currently do not show where the Family+Kingdom have changed – might need some more work to do this.
What are the implications for both possible users and data providers with this approach? What are the implications/roles of the web services and LSID resolution that can provide more than the components above within an external index or site? If the web service enables access to a more complete nomenclator record via a public interface, for example, or an LSID provides these details through its resolution, then when should these services be called, and when shouldn’t they?
Needs some thought on how GSD and Checklist data can ‘migrate’back down to the Nomenclators where there are identified conflicts (e.g. different orthography, different authors etc) – perhaps this could be managed by the Names Resolution Service in reverse – a name annotation service which the Nomenclators could then use to updata/augment content as appropriate.
Technology exists to facilitate this. How best implemented within a federated index of nomenclator/GSD/checklist data?
How should unverified or even verified names originating outside of the nomenclators (for example the BDWD fly genera) be incorporated & discoverable within for example, the AGI or a common index and subsequently made available to the nomenclators.
Need a workflow documented that details this process. How best to develop it?
Someone needs to get to grips with bibliography data; I want some standard lists of core systematics journals with dates of publication for each volume-part and standardized Full titles and Abbreviated forms – Roger Hyam tried to kickstart this last year – I’m sure most of this is already available – it just needs harvesting and working in to some useful form.
What is status of this? What would it require to shape into a useful service
Is something similar needed for author names? I know there are compilations of author abbreviations and uBio built a preliminary web service around these. Is a Author Resolution Service a practical ambition?
Comprehensive, consensus management classification – The plant and fungal nomenclators use a management classification. GBIF has a kludgy synthesized classification for occurrence data. AGI has a provisional one provided by Paddy. There are other quite practical applications for one (or should there be more than one?). Could the GNOMA include a common names management framework for easier disambiguation of inter-regnal names and enhanced interoperability? (both among nomenclators and with other data management initiatives like BHL/GBIF who could also employ such a management classification as one method for organizing names-annotated data). The issue arose at the PESI workshop. Should it go forward and how?
