Wikidata:Property proposal/iNaturalist observation ID

iNaturalist observation ID

Originally proposed at Wikidata:Property proposal/Natural science

Done: iNaturalist observation ID (P5683) (Talk and documentation)

Description	observation identifier in iNaturalist
Data type	External identifier
Domain	living things
Allowed values	[1-9]\d*
Example 1	Eutricha capensis (Q12347546) image (P18) value reference → 14692339
Example 2	Procavia capensis (Q323847) image (P18) value reference → 14693435
Example 3	Maniola jurtina (Q754668) image (P18) value reference → 14349142
Source	iNaturalist.org
Planned use	As reference for Wikidata statements
Formatter URL	https://www.inaturalist.org/observations/$1
See also	iNaturalist taxon ID (P3151)

Motivation

iNaturalist is a citizen science project that started in 2008. It is a widely used app that comes with a global community. The workflow is simple, observations (images, sounds, etc) are recorded and the community names and discusses the observation leading to research grade annotations. These annotations are valuable references for Wikidata statements. Andrawaag (talk) 03:33, 27 July 2018 (UTC)[reply]

Discussion

Support (with the caveat that we sill need to address the issues highlighted at Wikidata:Project chat#What heart rate does your name have?). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:06, 28 July 2018 (UTC)[reply]
Oppose I doubt this is within the scope of WD. They have no value for referencing our information. --Succu (talk) 20:33, 28 July 2018 (UTC)[reply]

Can you elaborate a bit on this? I have been using iNaturalist URLs in the past, so I don't see why it does not fit within the scope of WD. The problem with URLs, is that they are not persistent, which would vouch for a external ID property, which is why I am proposing it. --Andrawaag (talk) 16:49, 29 July 2018 (UTC)[reply]

@Andrawaag:

An observation is a special kind of occurence. GBIF records more than 1 Mrd. occurences! Around 5 Mio. are contributed by iNaturalist through a quality management process. That's what I meant with „I doubt this is within the scope of WD“
All of your picture examples could be used everywhere. Why should we favor the usage by iNaturalist? What about Commons?

--Succu (talk) 20:15, 1 August 2018 (UTC)[reply]

@Succu The purpose of this proposal is to store those pictures on Commons. iNaturalist is here not seen as a repository of images/observations, but as a community of curators. We have created a Wikiproject: Wikidata:WikiProject_iNaturalist to leverage the knowledge in iNaturalist to annotate images on commons. So the use case here is that a picture is uploaded to both commons and iNaturalist. When the observation is annotated on iNaturalist, this annotation is added to its identical picture on commons through Wikidata with the image property. It is here that a iNaturalist observation ID property is needed as a reference. So it is not the observation perse that is of interest, but the metadata of that observation. In the long run, it would be nice, to directly feed iNaturalist from commons. --Andrawaag (talk) 07:53, 2 August 2018 (UTC)[reply]

So the use case here is that a picture is uploaded to both commons and iNaturalist. Sorry this is a workflow, not a use case. I think WD should not be misused as a proxy for cleanups like this. --Succu (talk) 20:38, 17 August 2018 (UTC)[reply]

Yes I agree with you, the images on taxons lack credibility, of the 190k images to a taxon only about 1000 contain a reference. Adding pictures annotated by the iNaturalist community add credibility to those pictures, if a picture has a research grade ID on iNaturalist, that image will be of a higher quality than just any image. I don't think this is a cleanup, no it is unreasonable to expect the iNaturalist community to annotate the 190k images, but with this proposed property we can start feeding the taxon items in Wikidata with higher quality images (better annotations) than currently is the case. Now we just need to trust the submitter/photographer and as WD is not intended a primary source, we should not. --Andrawaag (talk) 06:07, 18 August 2018 (UTC)[reply]

Support. Thierry Caro (talk) 11:00, 29 July 2018 (UTC)[reply]
Comment Not sure what is the proposed usage, it is a qualifier of image (P18)? --Jklamo (talk) 11:48, 1 August 2018 (UTC)[reply]
- I still have not recieved answer, so Oppose (unclear usage). If it is inteded to be qualifier of image, Commons:Commons:Structured data is good place fot it, not Wikidata. It it is inteded as direct taxon property, it may clutter items with tens of observations for popular taxons.--Jklamo (talk) 19:14, 17 August 2018 (UTC)[reply]
  - @Jklamo I have missed your question, but tbo I don't understand the question since the answer is given in the request. See the examples. A lot of photo's are added to statements in Wikidata without proper referencing. iNaturalist provided the annotations of an image given. I honestly don't see how structured commons could solve that. Please read the examples given. Where reference URL is given, the proposed property would be a better fit. --Andrawaag (talk) 05:48, 18 August 2018 (UTC)[reply]