Open ingoboerner opened 3 years ago
I just saw that there is already a hint in <remarks>
at least in <castGroup>
that suggests using the attribute @corresp
. I am not sure if there is already support for this encoding or if we agreed to do it this way. maybe @cmil you can help me out with this:
<attDef ident="corresp" mode="change">
<datatype>
<dataRef key="teidata.pointer"/>
</datatype>
<remarks>
<p>Used to link a character in the dramatis personae to the
corresponding <!-- person/group? --> element in the
<gi>particDesc</gi></p>
</remarks>
</attDef>
Do you remember adding this @lehkost. It could have been me but I forgot ...
could include it in #67
I remember that we talked about this and @corresp
is definitely the way to go. I would not use this additional encoding for the corpora that I'm working on right now as there is no direct use (info in particDesc usually is enough to do network analysis, etc.). But we should allow (and even recommend) this to establish a semantic bridge between particDesc and castList.
Currently, in DraCor in-house corpora we include the
tei:castItem
in atei:castList
that contains the name of a character and his/her roleDescription ... in addition, to link speech acts to speakers/characters, the files contain atei:listPerson
withtei:person
that are identified by an@xml:id
. ButcastItem
and corresponding person are not explicitly linked (via an attribute). If one want's to get the "label" of a character as contained in the text, this can only be done via complicated string comparisons and fuzzy matching (?), becausecharacters and correspondingtei:person
are not in the same order, so the link between these two elements is not even somehow implicit. Maybe, it should be made explicit (@corresp
) to provide for an easier way to retrieve the information from thetei:castItem
via the@xml:id
of thetei:person
in thetei:particDesc
Example of current encoding: goethe-egmont
and