[Taxacom] Species Cite: linking scientific names to publications and taxonomists

[Roderic Page]

Hi Tony,

Yes, one could argue that this is exactly what we could be doing, outsourcing taxonomic names and publications to Wikidata where there is a community of people with diverse interests activity editing and enhancing the data.

And outsourcing need not mean complete dependence, one could have a local database with your own records and simply store a mapping to Wikidata and take from Wikidata whatever information is useful.

I think we are likely to see Wikidata become the centralised taxonomic name database by default, given that it is probably richer and more actively edited than anything our community produces. That said, because it doesn’t cleanly separate taxonomic names and taxa things are messy and inconsistent, and likely to remain so. I think at it’s heart there is a fundamental mismatch between what people want (a database of things, i.e., actual species) and what it is (a collection of names and multiple, half-formed classifications).

For now I’ve found it easier to focus on the literature because (a) it’s a big gap in our knowledge and (b) it is shared across taxonomy (multiple different databases may cite the same article) and there is less sense ownership compared to taxonomic names.

So I guess the short answer to

Then, why not just put the taxon names into Wikidata as well, with links to the references in which they were originally published... I am guessing some of those are there already. Then all you would need would be to build some taxon name viewer to sit over the top.

is that this is sort of happening in a somewhat messy way. Personally I’m trying to figure out whether to embrace the mess. Braver souls have already jumped in.

Regards,

Rod

On 3 Aug 2021, at 07:40, Tony Rees <tonyrees49 at gmail.com<mailto:tonyrees49 at gmail.com>> wrote:

Hi Rod,

Not answering my own questions as posed above, but on a more philosophical note... just wondering what you see as an "end point" for this object (data) linking exercise...

You talk about having all the references in Wikidata as things you (and anyone else) can then link to. Of course, one could envisage a model whereby WoRMS or IRMNG could also port their references to Wikidata at some point - assuming they are not already there - then call those up as needed to display in taxon pages. Enter once, use many times, a good principle. (Also would enable e.g. WoRMS and IRMNG to avoid making duplicate entries for the same reference, as is currently the case).

Then, why not just put the taxon names into Wikidata as well, with links to the references in which they were originally published... I am guessing some of those are there already. Then all you would need would be to build some taxon name viewer to sit over the top.

Then, the Wikidata references could reference Wikidata entries for each contributing author, and those could reference an Orcid (at least for the living ones, not sure about the non-living), giving the "persons" element ... provided that these were populated of course.

So in essence I am saying, can you not just outsource all the content and links between data items, for example to Wikidata in this instance, let the "community" maintain them there, then just build a nice tool to sit over the top and display what is needed in answer to user queries? If this means porting the items and links you already have in your own system to there, is that not a good thing?

Just wondering,

Best - Tony


On Tue, 3 Aug 2021 at 16:26, Roderic Page via Taxacom <taxacom at mailman.nhm.ku.edu<mailto:taxacom at mailman.nhm.ku.edu>> wrote:
Hi Geoff,

Apologies for the mischaracterisation, you can probably tell I’m not a major WoRMS user, although I do occasionally spend time there tracking down names of freshwater and terrestrial molluscs.

Yes, WoRMS includes DOIs and links. I don’t have a full copy of the WoRMS references, but based on what I have (~150,000) around 10% of references have a DOI, and less than 5% are linked to BHL.

These numbers give us a sense of the scale of the task, given that WoRMS has an active community of editors (I’m guessing WoRMS might claim to have the largest community of editors among taxonomic databases?). So much of the biodiversity literature across our databases remains unconnected and undigitised, and the work of making those connections is not fun (although strangely satisfying).

I’m focussing my editing efforts on Wikidata. I note that Wikidata has a property for WoRMS sources which enables references in Wikidata to be linked to WoRMS, so there is a mechanism in place to link records in the two projects. My expectation is that these links will uncover information in both sources that the other wasn’t aware of.

Regards,

Rod