search

plazi / arcadia-project

2 stars 1 forks source link

Audubon core: use in BLR metadata? #137

Open myrmoteras opened 4 years ago

myrmoteras commented 4 years ago

Dear Terry and Marcus steve Baskauf made an interesting suggestion to use terms from the audubon core to describe the content of images we deposit at BLR. For example we could add an element to describe the view: lateral, dorsal, full frontal, or what is on the image such as s leg, head, etc. May be this is something to suggest Fabian for the Fabricius types? May be Steve, you could provide a link to the new release, a comment or write a comment for us at Plazi as a news item. This is something we have been struggling with during the ICEDIG project, and this is an argument in our discussion on copyright why scientific illustrations are not works in a legal sense. In fact we probably could easily make such annotations in BLR. See eg https://ocellus.punkish.org/images.html?communities=biosyslit&q=Lateral&size=30&page=1 gor figures containing a leteral view. Just a thought Donat

myrmoteras commented 4 years ago

I think this is a good idea. However, to be clear, AC provides terms like "subjectOrientation" and "subjectPart" not "dorsal" or "head" for which, as the documentation states: "No formal encoding scheme as yet exists." So we'd have to use whatever is present in an associated caption, normalize and QC it , and perhaps assign our own controlled terms.

Also, ac:otherScientificName could be very useful for dealing with synonyms.

Thanks for the ideas,

Terry

myrmoteras commented 4 years ago

I'm just getting caught up on email from getting back from Poland so will make a brief reply. I'm also moving this thread over from my personal gmail to my Vanderbilt email.

Right now there is a task group that is working on developing the controlled vocabularies for ac:subjectOrientation and ac:subjectPart. See https://github.com/tdwg/ac/tree/master/views . We've solicited use cases (https://github.com/tdwg/ac/blob/master/views/submitted-use-cases.md) and are just getting started on actually developing the controlled vocabulary terms. So it would be extremely timely to have some feedback from you about how you would use this vocabulary and any data that you have about subject orientations and parts that are actually being asserted in the wild. Our goal is to develop controlled vocabularies that will actually be used, so having communication with potential users would be extremely valuable to the task group.

The most recent version of Audubon Core is here: https://tdwg.github.io/ac/termlist/ in human-readable form. If you are interested, I can let you know how to get the data in machine-readable form.

Steve Baskauf, Steven James steve.baskauf@Vanderbilt.Edu