Hi Rod.

Given the two alternatives you offer, I fully agree. However (and I may be very wrong on this), I don’t actually believe that what you describe is what is happening in most cases. I believe that the scientificName field should always be the “give us the cleanest representation of the scientific name you have” and we could certainly provide a more useful set of priorities for how that is defined. Regardless of what the DwC guidelines say, I think many data providers simply map their most closely aligned database column into the DwC view and we get whatever that happens to be, with whatever authorship it may contain.

I’d agree that the clean name is much closer to what a web-enabled linked-data world needs and would happily endorse a move to make that the recommended form. I just honestly believe that many providers will always give us something different. Adding a new field will just shift the problem.

At least that’s my perspective

All the best,

Donald

----------------------------------------------------------------------

Donald Hobern - GBIF Director - mailto:dhobern@gbif.org dhobern@gbif.org

Global Biodiversity Information Facility http://www.gbif.org/ http://www.gbif.org/

GBIF Secretariat, Universitetsparken 15, DK-2100 Copenhagen Ø, Denmark

Tel: +45 3532 1471 Mob: +45 2875 1471 Fax: +45 2875 1480

----------------------------------------------------------------------

From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Roderic Page Sent: Wednesday, March 14, 2012 5:50 PM To: TDWG Content Mailing List Subject: Re: [tdwg-content] canonicalScientificName

Dear Donald,

I couldn't disagree more!

It seems to me that this is one case where needs of consumers and providers align pretty well.

If I'm publishing data I want to avoid hassle, and one hassle is finding the taxonomic authorities for names. Then there is the issue of how to write the authority. There are so many variables: do I include diacritic characters? is the person's name abbreviated? what is the correct date? should I use parentheses? should I use commas? If I can just publish the canonical name life is simpler.

As a consumer I can't trust people to get the authority right. Publishers get the taxonomic names wrong, and they will certainly make a mess of the authority.

So, if we mandate clean names we are saying to providers "give me this"

<taxonomic name>

[some scope for crap]

Instead, we've mandated "give me this"

<taxonomic name> + <authority>

[some scope for crap] + [huge scope for crap]

Why? Why would we do this to ourselves? Why do we think it's OK to have databases full of duplicates such as these (from the ION database)?:

Pseudopaludicola Miranda Ribeiro 1926

Pseudopaludicola Mir. Ribeiro 1926

Pseudopaludicola Miranda-Ribeiro 1926

One consequence of this is that we have projects like http://globalnames.org project, which is essentially collecting endless variations on authority strings. In other words, trying to clean up a mess essentially of our own making.

By all means have a field for taxonomic authority, but keep that separate from the canonical taxonomic name. In the real world, the canonical name is what people use. If we want people to make data available, make it simple. If we want people to use data make it simple.

Regards

Rod

On 14 Mar 2012, at 11:16, Donald Hobern wrote:

Hi Peter.

I certainly sympathise with the desire for a readily-consumed naked scientific name field. However, unless the canonicalScientificName element is enforced as a mandatory field (which would in itself impact some data publishers and may prevent them validly sharing their data without extra work to provide clean scientific names), it will be yet another element which data consumers must check. If canonicalScientificName is supplied, consumers will still need to handle cases where it is malformed. If is not supplied, they will need to ignore the record or else do precisely what they do today with the scientificName field.

I therefore worry that adding this field could in fact make the task more complex, rather than simpler, for data consumers.

Thanks,

Donald

----------------------------------------------------------------------

Donald Hobern - GBIF Director - mailto:dhobern@gbif.org dhobern@gbif.org

Global Biodiversity Information Facility http://www.gbif.org/ http://www.gbif.org/

GBIF Secretariat, Universitetsparken 15, DK-2100 Copenhagen Ø, Denmark

Tel: tel:%2B45%203532%201471 +45 3532 1471 Mob: tel:%2B45%202875%201471 +45 2875 1471 Fax: tel:%2B45%202875%201480 +45 2875 1480

----------------------------------------------------------------------

_______________________________________________ tdwg-content mailing list tdwg-content@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-content

--------------------------------------------------------- Roderic Page Professor of Taxonomy Institute of Biodiversity, Animal Health and Comparative Medicine College of Medical, Veterinary and Life Sciences Graham Kerr Building University of Glasgow Glasgow G12 8QQ, UK

Email: r.page@bio.gla.ac.uk Tel: +44 141 330 4778 Fax: +44 141 330 2792

Skype: rdmpage AIM: rodpage1962@aim.com Facebook: http://www.facebook.com/profile.php?id=1112517192 Twitter: http://twitter.com/rdmpage Blog: http://iphylo.blogspot.com Home page: http://taxonomy.zoology.gla.ac.uk/rod/rod.html

Hi Peter.

I should stop commenting because I’m only one opinion here but, in response to your question, I think that in practice we will end up with two fields, each of which is treated in a rather lax fashion by most providers, so we end up still having to do as much work to clean data, but now with the added uncertainty of which field to use if the two are different.

As I indicated, given that scientificName is the existing mandatory field, I’d be more in favour of reviewing whether our current definition for that field is in fact best. Encouraging everyone who can give a simple easily-processed name to do so in the scientificName field and continuing to use that as the mandatory species identifier would simplify name matching for most users. We should be relying on nomenclatural databases for the author citation details. In those cases where we could use author names to disambiguate homonyms we can usually (I realise not always) use higher taxonomy much more easily for the same purpose.

I agree heartily with the goal behind the suggestion. I however doubt that it will achieve the critical mass needed to enable consumers to rely on it. To summarise, my preferred strategy would be to use one or all of the following:

1. Change the DwC scientificName definition to recommend the use of a clean monomial, binomial or trinomial (with whatever seem appropriate recommendations for varieties and lower ranks) rather than the current fully-formatted form

2. Place an obligation on GBIF and other aggregators to provide robust, predictable services to access “clean” versions of these data sets

3. (A variation on 2) Develop services that can be used in a unix-pipe fashion to take any DwC-A file and produce a clean version

Best wishes,

Donald

----------------------------------------------------------------------

Donald Hobern - GBIF Director - mailto:dhobern@gbif.org dhobern@gbif.org

Global Biodiversity Information Facility http://www.gbif.org/ http://www.gbif.org/

GBIF Secretariat, Universitetsparken 15, DK-2100 Copenhagen Ø, Denmark

Tel: +45 3532 1471 Mob: +45 2875 1471 Fax: +45 2875 1480

----------------------------------------------------------------------

From: peter.desmet.cubc@gmail.com [mailto:peter.desmet.cubc@gmail.com] On Behalf Of Peter Desmet Sent: Wednesday, March 14, 2012 7:04 PM To: Donald Hobern (GBIF) Cc: Roderic Page; TDWG Content Mailing List; TDWG TAG mailing list Subject: Re: [tdwg-content] canonicalScientificName

Hi Donald,

I just don't understand why an additional field canonicalScientificName (much like minimumElevationInMeters or countryCode) would create or shift the problem. It has a clear and easy to understand definition [1]. As a data provider I can assess if its too much of a bother to populate the field in addition to the more lenient scientificName (just like countryCode, etc.) and as a data user I can complain or ignore the data if I see the provider didn't follow the definition.

This discussion resurfaces every time: clearly users would like to have this. So why do we make it difficult for those users and the data providers who can provide it?

Peter

[1] http://code.google.com/p/darwincore/issues/detail?id=150 I could even add "[...] it will contain one, two or three words" (see email Jessie Kennedy)

PS: Yes, some people probably interpret scientificName as "Give us the cleanest representation of the scientific name you have", but that is definitely not the Darwin Core definition. It's more "give us the most verbose representation of the scientific name you have", which is what I use and what I advise all data publishers in my network to use. The difference doesn't matter though, the important thing is that a data user cannot expect this term to be a canonical representation of the name.

On Wed, Mar 14, 2012 at 13:36, Donald Hobern (GBIF) dhobern@gbif.org wrote:

Hi Rod.

Given the two alternatives you offer, I fully agree. However (and I may be very wrong on this), I don’t actually believe that what you describe is what is happening in most cases. I believe that the scientificName field should always be the “give us the cleanest representation of the scientific name you have” and we could certainly provide a more useful set of priorities for how that is defined. Regardless of what the DwC guidelines say, I think many data providers simply map their most closely aligned database column into the DwC view and we get whatever that happens to be, with whatever authorship it may contain.

I’d agree that the clean name is much closer to what a web-enabled linked-data world needs and would happily endorse a move to make that the recommended form. I just honestly believe that many providers will always give us something different. Adding a new field will just shift the problem.

At least that’s my perspective…

All the best,

Donald

----------------------------------------------------------------------

Donald Hobern - GBIF Director - dhobern@gbif.org

Global Biodiversity Information Facility http://www.gbif.org/

GBIF Secretariat, Universitetsparken 15, DK-2100 Copenhagen Ø, Denmark

Tel: +45 3532 1471 tel:%2B45%203532%201471 Mob: +45 2875 1471 tel:%2B45%202875%201471 Fax: +45 2875 1480 tel:%2B45%202875%201480

----------------------------------------------------------------------

From: tdwg-content-bounces@lists.tdwg.org [mailto:tdwg-content-bounces@lists.tdwg.org] On Behalf Of Roderic Page Sent: Wednesday, March 14, 2012 5:50 PM To: TDWG Content Mailing List Subject: Re: [tdwg-content] canonicalScientificName

Dear Donald,

I couldn't disagree more!

It seems to me that this is one case where needs of consumers and providers align pretty well.

If I'm publishing data I want to avoid hassle, and one hassle is finding the taxonomic authorities for names. Then there is the issue of how to write the authority. There are so many variables: do I include diacritic characters? is the person's name abbreviated? what is the correct date? should I use parentheses? should I use commas? If I can just publish the canonical name life is simpler.

As a consumer I can't trust people to get the authority right. Publishers get the taxonomic names wrong, and they will certainly make a mess of the authority.

So, if we mandate clean names we are saying to providers "give me this"

<taxonomic name>

[some scope for crap]

Instead, we've mandated "give me this"

<taxonomic name> + <authority>

[some scope for crap] + [huge scope for crap]

Why? Why would we do this to ourselves? Why do we think it's OK to have databases full of duplicates such as these (from the ION database)?:

Pseudopaludicola Miranda Ribeiro 1926

Pseudopaludicola Mir. Ribeiro 1926

Pseudopaludicola Miranda-Ribeiro 1926

One consequence of this is that we have projects like http://globalnames.org project, which is essentially collecting endless variations on authority strings. In other words, trying to clean up a mess essentially of our own making.

By all means have a field for taxonomic authority, but keep that separate from the canonical taxonomic name. In the real world, the canonical name is what people use. If we want people to make data available, make it simple. If we want people to use data make it simple.

Regards

Rod

On 14 Mar 2012, at 11:16, Donald Hobern wrote:

Hi Peter.

I certainly sympathise with the desire for a readily-consumed naked scientific name field. However, unless the canonicalScientificName element is enforced as a mandatory field (which would in itself impact some data publishers and may prevent them validly sharing their data without extra work to provide clean scientific names), it will be yet another element which data consumers must check. If canonicalScientificName is supplied, consumers will still need to handle cases where it is malformed. If is not supplied, they will need to ignore the record or else do precisely what they do today with the scientificName field.

I therefore worry that adding this field could in fact make the task more complex, rather than simpler, for data consumers.

Thanks,

Donald

----------------------------------------------------------------------

Donald Hobern - GBIF Director - mailto:dhobern@gbif.org dhobern@gbif.org

Global Biodiversity Information Facility http://www.gbif.org/ http://www.gbif.org/

GBIF Secretariat, Universitetsparken 15, DK-2100 Copenhagen Ø, Denmark

Tel: tel:%2B45%203532%201471 +45 3532 1471 Mob: tel:%2B45%202875%201471 +45 2875 1471 Fax: tel:%2B45%202875%201480 +45 2875 1480

----------------------------------------------------------------------

_______________________________________________ tdwg-content mailing list tdwg-content@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-content

--------------------------------------------------------- Roderic Page Professor of Taxonomy Institute of Biodiversity, Animal Health and Comparative Medicine College of Medical, Veterinary and Life Sciences Graham Kerr Building University of Glasgow Glasgow G12 8QQ, UK

Email: r.page@bio.gla.ac.uk Tel: +44 141 330 4778 tel:%2B44%20141%20330%204778 Fax: +44 141 330 2792 tel:%2B44%20141%20330%202792

Skype: rdmpage AIM: rodpage1962@aim.com Facebook: http://www.facebook.com/profile.php?id=1112517192 Twitter: http://twitter.com/rdmpage Blog: http://iphylo.blogspot.com Home page: http://taxonomy.zoology.gla.ac.uk/rod/rod.html

_______________________________________________ tdwg-content mailing list tdwg-content@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/tdwg-content