Like you, I've never really liked the label-followed-by-item structure;
I don't like relationships based on contiguity, because they end up
being ambiguous. For instance, in this:
Does Label 2 apply to Text X, along with Label 3, or is it indicative of
a missing text?
I think this should be the recommended structure:
<p> Text 1 </p>
(where <p> of course could be any other appropriate element.
In P6, the contiguity option should not be valid. (Cue arguments about
P6, when, whether, why, what, etc.)
On 2019-01-26 6:43 p.m., Martin Mueller wrote:
> What is the difference, if any, between <item><label/></item and <label/></tem/> ? I'm aware of the tradition of the "glossary", but from a conceptual perspective a label relates to an item pretty much in the same way in which <head> relates to <div>. If that is the case, why treat lists differently from divs?
> Somebody could argue that the <label/><item/> structure is more like a key:value pair. But there is no constraint on repeated labels. A list can have a <head> element that names it, just as the label names the item. Wouldn't it be simpler to maintain the same structure recursively?
> This is main, but not entirely, an idle question. I've been wondering about the relationship between <head> and <label> and about heads or labels in paragraphs. Paragraph heading are quite common in certain forms of prose.
> If you recognize a function in which some discursive structure is accompanied by shorter element that tells you what it is about, do you help or hinder the recognition of that function by having two different elements?
> Martin Mueller
> Professor emeritus of English and Classics
> Northwestern University