University of Glasgow

Roderic Page

Quick notes on taxonomic names (again). It\'s a continuing source of
bafflement that the biodiversity community is making a dog\'s breakfast
of names. It seems we are forever making it more complicated than it
needs to be, forever minting new acronyms that pollute the landscape
without actually contributing anything useful, and forever promising
shiny new tools and services without every actually delivering them.
Meanwhile people and projects that build upon names are left to deal
with a mess.  
  
It seems to me that it would be nice if we had a single place to go to
get definitive information on a name, and that place would give us a
unique identifier that we could use in our own databases as a way to
clean up and reconcile our data. For example, if we have a bibliographic
database we can map citations to DOIs and then use those to identify the
articles. If we have a list of journal names, we can map those to ISSNs
and clean up our data. Likewise, if we have a classification such as
GBIF or NCBI, we should be able to map the names in those
classifications onto standard identifiers for taxonomic names.  
  
The frustrating thing is we **already have standard identifiers for
taxonomic names**. Since around 2005 we have been serving LSIDs for
plant and animal names. We have [Index
Fungorum](http://www.indexfungorum.org), [IPNI](http://www.ipni.org),
[ION](http://www.organismnames.com), and [ZooBank](http://zoobank.org),
all serving LSIDs, all serving RDF, all using the **same** TDWG
vocabulary.  
  
The nomenclators vary in size and scope, but we have the three major,
multicellular eukaryotes covered (circles proportional to number of
names in each database):  

::: {style="text-align:center"}
![Nomenclators](http://lh5.ggpht.com/-vNQaFASi6tU/UoDy7-qp1cI/AAAAAAAADLE/W5ilYEbO3DM/nomenclators.png?imgmax=800 "nomenclators.png"){border="0"
width="450" height="242"}
:::

  
There is some duplication, both within nomenclators (IPNI and ION I\'m
looking at you) and between nomenclators (ION and ZooBank have the same
scope, although ZooBank is dwarfed by ION, anyone care to explain why we
have both\...?). All four databases are actively growing, partly through
direct registration of new taxonomic names.  
  
So, we\'re basically done, right? Surely all we need to do is harvest
the LSIDs for all these names, put them into a single triple store, and
wrap some basic services around them? If the nomenclators provide a list
of recent changes (e.g., as an RSS feed) then we could continuously
update the store with new names. Then any database or classification
could reconcile it\'s names with those in the nomenclators. They could
also then augment their own records by making use of additional data the
nomenclators have, such as objective synonomies and links to original
descriptions. In other words, we could have a model like this:  
![Taxonmodel](http://lh5.ggpht.com/-IdD1vHyXoNA/UoDy8-htQCI/AAAAAAAADLM/BESVrG4kC-c/taxonmodel.png?imgmax=800 "taxonmodel.png"){style="display:block; margin-left:auto; margin-right:auto;"
border="0" width="450"}  
Classifications represent a view of how taxa are related, the names
associated with those taxa are stored in nomenclators. This means that
classification databases like GBIF and NCBI are not in the business of
managing names, they simply link to the nomenclators (in the same way
that a bibliographic database can link to DOI, ISSNs, and author ids
such as ORCID and VIAF).  
  
We have almost all of this infrastructure in place already. In one of
the unsung triumphs of [TDWG](tdwg.org) we have all the nomenclators
serving data in the same format using the same technology. And yet we
have singly failed to do anything useful with this extraordinary
resource! Instead we seem more interested in contributing more projects
to the acronym soup of biodiversity informatics. All around us projects
to assign and link identifiers for publications (CrossRef), data
(DataCite), and people (ORCID) are taking off. The infrastructure for
taxonomic names has been in place since 2005, we could be doing the same
sort of things CrossRef, DataCite and ORCID are doing in their domains.
Why aren\'t we?


Quick notes on taxonomic names (again). It's a continuing source of bafflement that the biodiversity community is making a dog's breakfast of names. It seems we are forever making it more complicated than it needs to be, forever minting new acronyms that pollute the landscape without actually contributing anything useful, and forever promising shiny new tools and services without every actually delivering them.

Names and nomenclators: just do it already!

![DD01286D-DB9A-46B6-A05B-5B6A42CE0747.jpg](http://lh6.ggpht.com/_Gct8lVAxKqQ/SmkPaSp7zZI/AAAAAAAAAgg/HZly56eDCmY/DD01286D-DB9A-46B6-A05B-5B6A42CE0747.jpg?imgmax=800){border="0"
width="100" height="100" align="right"}I\'ve added [Index
Fungorum](http://www.indexfungorum.org/Names/Names.asp) to the list of
RSS feeds that I generate at
[bioguid.info/rss](http://bioguid.info/rss). The feed uses the Index
Fungorum web services to get the names added the previous day, and tries
to extract any bibliographic identifiers from the metadata associated
with each record (we get the metadata by resolving the LSID for the
name). [As with
IPNI](http://iphylo.blogspot.com/2009/05/nomenclators-digitised-literature-fail.html),
bibliographic information in an Index Fungorum record lists the page the
name was published on, which makes locating identifiers such as DOIs a
bit of a struggle. Still, it\'s nice to have another feed of taxonomic
names.


I've added Index Fungorum to the list of RSS feeds that I generate at bioguid.info/rss.

Messaggi di Rogue Scholar

Names and nomenclators: just do it already!

Index Fungorum