<said>, <q>: att classes and behavior in context of Overlapping Hierarchies

[ebeshero]

The Guidelines would benefit from more examples and advice on how to re-model the contents of spoken passages (via <said> and <q>, etc) in the contexts of overlapping hierarchies, since this is a frequent concern for coders of verse structures.

Questions: One way to model an overlapping structure is by working with self-closing elements set in sequence, using attributes to demonstrate how these connect. We currently have no guidance on how this could or should not work for <said> or <q> in the Guidelines--either to advise for it or against it. What meanings can a self-closed <said/> indicate, with or without spanning or connecting attributes present?

• Can it indicate that something was expected to be said but wasn't? See the scenario described on the TEI listserv by @martindholmes here for a comic book panel with an empty speech bubble.

• Can it be used in contexts of overlapping hierarchy to mark the start of a sequence, when <said> and its companions <q> and <quote>? If so, can these be added to att.spanning (specifically defined for this purpose)? See related ticket: https://github.com/TEIC/TEI/issues/197

• Better yet, shouldn't these elements be added to att.fragmentable? http://www.tei-c.org/release/doc/tei-p5-doc/en/html/ref-att.fragmentable.html This would provide a more nuanced treatment of interrupted passages of speech, as in: "I already took the dog for a walk," she said, "but he wants to go out again."

[raffazizzi]

@ebeshero would you like to revisit this and narrow it down to a proposal? Or should we consider closing it?

[JanelleJenstad]

@JanelleJenstad would like to join forces with @ebeshero to develop a proposal.

[ebeshero]

We should consider adding a standoff example to this as well.