NPL's Darwin Core mappings

In order to create the data sets required by VertNet, we have to have functional schema mappings. This is basically a translation from our field names to those of Darwin Core. Darwin core terms are in the drop down to the left, and the Specify fields are to the right. Attached is a xml file of these mappings. If you want to use this file, simply download it, and in the Specify Schema Mapper Interface, right click on the Mappings header and select 'import' from the pop up window.

 

The file can be downloaded by clicking on the link:

Sections in this Chapter:


 

 Starting a new mapping

First, navigate to the Schema Mapping interface by clicking on the 'System' label at the top of the Specify interface.

In the side bar, select 'create new mapping'. You'll be prompted to choose an export schema. We use DarwinCoreTerms2.

When the interface opens, many things will already be mapped, but they are in no particular order. At this point, I usually take a few moments to organize the fields according to table.

 

As your organizing your interface, I'm sure you'll notice things that are not really necessary. For some reason, auto mapping thinks our dataset readers want to know what day the database record was modified on. Keep an eye our for things like that, often times they are not correctly mapped, or just plain irrelevant. Clicking on the X removes the line.

Now, to add a line to the interface simply follow the same steps you'd use to add a field to a query interface. Fish it out of the upper part of the interface, the one that shows you the tables and their contents, and click on the field once you've found it. Of course, it will be added at the bottom, so you'll have to move it up to group it with the others of its table. Here, you see we've added the Field Number to the interface, and moved it up so that it's grouped with the rest of the Collecting Event fields.

 

Once you've successfully created a mapping, let the Collection Manager know and she can compile the data set and export it to our host. If you are instructed to use the Data Exporter, which requires high level log-in clearance, you can find the program in C:\Program Files (x86)\Specify\bin.

Creating a mapping from scratch takes a lot of time, but it's a good learning experience.


 

 Tips and tricks

You will come to hate this error message:

 

What this means is there is an issue with your mapping- somewhere, there is a duplicate entry (and by duplicate, I mean even though they will be totally different records, there is a grain of data that is the same between the two of them).

 

 

The best way to combat this is to aggregate the table by double clicking on it. Notice the entry in the schema display- it has the word 'Formatted' next to the field that was double clicked.

 

 

 

 

 

 

Common trouble makers who need this treatment are: agents (in any role- determiner, author, etc..,) Inventory and Preps, and GeoRefDetail.

 

The other issue that causes this error message is Determination. You MUST add the checkbox 'Current' to the mapping in order to work around this. It also has to be set to 'Yes' for it to work.

 

Error messages, how and how to work around the most common ones. Add your own experiences in the comments!