[tei-council] said at part

[Syd Bauman]

> Going through NH, I see that said at part is not allowed. @part shows
> up in att.segLike
> (http://tei.oucs.ox.ac.uk/P5/Guidelines-web/en/html/ref-att.segLike.html)
> and ab
> http://tei.oucs.ox.ac.uk/P5/Guidelines-web/en/html/ref-ab.html Does
> anybody know why this is neither global nor restricted solely to
> segs?

Well, I don't really know why, but part= is defined separately for
att.divLike, att.segLike, and <l>. (IIRC, this has it on the same
elements as in P4.) This attribute has only limited usefulness when
elements can self-nest. So for <l> and <div2> it's very useful. But
for <s>, <seg>, <said> (and even <div>), I prefer to avoid using it,
because it's easy to encode things ambiguously:
<p>a b <seg part="I">c d <seg part="I">e f</seg> g h</seg> i j</p>
<p>k l <seg part="F">m n <seg part="F">o p</seg> q r</seg> s t</p>

You can't make that mistake with HORSE or next=/prev=. (You can, of
course, still point to the wrong element, but that's a different kind
of mistake :-)

So while I can see an argument for adding it to <q>, <quote>, and
<said>, the above argument suggests not doing so, and in fact
removing it from <cl>, <m>, <phr>, <s>, <seg>, <w>, <div>, and <lg>.