Inconsistencies in "connections" to external concepts

[ingoboerner]

We have several encoding strategies in place to link out to external concept:

<author>
          <persName>
           <!-- ... -->
          </persName>
          <idno type="wikidata">Q113227</idno>
</author>

vs.

<person xml:id="christus" sex="MALE" ana="http://www.wikidata.org/entity/Q302">
            <persName>Christus</persName>
</person>

vs.

<listRelation>
      <relation name="wikidata" active="https://dracor.org/entity/ger000171" passive="http://www.wikidata.org/entity/Q51370104"/>
 </listRelation>

vs.

<textClass>
        <keywords>
          <term type="genreTitle">Comedy</term>
        </keywords>
        <classCode scheme="http://www.wikidata.org/entity/">Q40831</classCode>
      </textClass>

On using @ana see also #71

[lehkost]

Thanks for the overview, @ingoboerner! On using @ana in person elements within particDesc, I remember this was a quick solution we found at the hackathon in Winter 2019 (or a year earlier?). It helped us to collect a lot of useful information on mythological and historic characters and even yielded its own API function (/character/{id}). If we can encode this info more consistently, I'm game. I just haven't found (or looked for) a better solution yet, also lacking examples from other projects.