[MEI-L] Semantically unspecific text (sub)divisions

Thomas Weber tw at notabit.eu
Wed Sep 30 11:37:40 CEST 2015

As far as I understand, in TEI, the elements <anchor>, <ab> and <seg> would be options to (sub-)divide/mark up a stretch of text without necessarily inducing some pre-defined semantic implications[1].  I see no way of doing that in MEI, e.g. for

* encoding some text that neither classifies as anchored text, heading, table, paragraph, line group, list or block quote.

* annotating a span of text inside a <p> or <anchoredText> element, but not the entire element.

Unless I'm missing something, would it be sensible to take the mentioned TEI elements as a model for MEI?  To the TEI experts - what exactly is the difference between <ab> and <seg>?  It seems to me <ab> is more like an HTML <div> (a "block" - is that what TEI calls a "chunk"?), while <seg> is more like an HTML <span> ("inline" in HTML terms).

Specifically I was looking into how to sensibly encode rubrics in neumatic notation, wondering whether I should use something like

<div type="rubric">
    <p>some rubric text</p>

but I don't think describing rubrics as paragraphs is semantically appropriate.  <head> could be considered appropriate in some cases, but it's not allowed as the only element inside a <div>.  I'll probably circumvent that problem by adding a NeumesXML-like <rubric> element[2] in our customization, but I feel the general problem that standard MEI might currently force one to classify text in potentially inappropriate ways remains.

One option could also be to allow text immediately inside <div> without having to wrap it into a more semantically specific element, but enforcing more semantical markup is of course not a bad idea, provided semantically appropriate choices are made available, including a fallback to more generic markup.

[1] http://www.tei-c.org/release/doc/tei-p5-doc/en/html/SA.html#SASE
[2] http://www.scribeserver.com/NEUMES/neumesxml_tree/tree_detail_west/rubric.htm

Thomas Weber
Burgkstra├če 28
01159 Dresden

Tel.: +49 (0)351 4794689

More information about the mei-l mailing list