[Tagging] Feature Proposal - RFC - Find A Grave cemetery and grave IDs

stevea steveaOSM at softworkers.com
Sun Mar 14 22:19:45 UTC 2021


On Mar 14, 2021, at 2:58 PM, Graeme Fitzpatrick <graemefitz1 at gmail.com> wrote:
> I've never understood Wikidata & what it's all supposed to do?

If I may take a stab at this, even at the risk of getting it wrong.  (And I have never really used Wikidata, only "watched" as it has evolved around me and what I do).

Wikidata seems to want to be the world's premier "knowledge graph" (its own description of itself) of "things."  Yes, it is multilingual and collaborative, like other Wikimedia Foundation - sponsored projects, so that's now said.

If you know the word "ontology," great.  If you don't, I'll include its relevant dictionary definition as a noun here and now:  "a set of concepts and categories in a subject area or domain that shows their properties and the relations between them."  Every single Wikidata "tagged item" ostensibly wishes to be the "canonical member of the ontology of things" (on Earth, as these are unique and identifiable to a particular noun, like the Eiffel Tower or Ayers Rock — and maybe even "off Earth," I haven't checked that).  There are associated labels, properties, descriptions, references — all the things that a "thing" that is useful to be able to have a "handle" to it should reasonably be expected to have.

What this does (again, ostensibly) is provide a framework of a "definitive ontology" that contains a full (or approaching it) knowledge graph of "things."  You can see how overlapping "things" in OSM with "things" in Wikidata can be helpful, as it allows a one-to-one mapping between an "OSM object" with "the unique thing in the real world" — importantly, UNAMBIGUOUSLY.

> Is it something that we as mappers should always be actively trying to link to, or is it just another If you feel like it, then do it?

I have "watched" Wikidata, as I say.  Its "uptake" has been swift, with much of how it "gets used" happening behind the scenes and under the covers of other projects, though as I understand what I've just described, that's only a good thing.  I feel no compunction to update, QA, improve or otherwise link to or "purposefully do something with" Wikidata tags, except to more-or-less be aware of what they are and how they seem to function.  So I describe how I think of them here, again at the expense of perhaps having others here knock off some of my rough edges of understanding, if necessary.  But I think I've largely got it right when I think of Wikidata (in the OSM context) as "an ontology, one which helps prevent ambiguity between things represented in OSM and the actual things that OSM is representing in the real world."

I hope that helps.  Again, I welcome corrections to my understanding.


More information about the Tagging mailing list