Lee wrote:
Jon Noring <jon@noring.name> wrote:
Maybe this time around the TEI mavens will be able to cogently explain (at the highest abstract level) why one should not have a <p> after a <div> within another <div>, but can have a <p> before the <div>. I'm still quite perplexed, as it appears Marcello is as well.
OK, so I guess I really don't want to know that badly. Everyone has to select from among those things they feel passsionate about, which ones they are going to spend time working one, and for me this is not one of those things. Besides, if they're not going to listen to Mr. Perathoner, or Mr. Collin, or even, apparently, Mr. Rahtz, one more random voice crying from the wilderness is not going to make much difference.
Yes, definitely. Your comment reminds me of the Book of Ecclesiastes in the Bible about the "right time" and "wrong time" to pursue something. Nevertheless, those working on TEI should subscribe to TEI-L and casually follow the discussion threads. It's quite interesting.
Besides, being the anarchist that I am, my solution would be just to grab the TEI dtds from tei-c.org, rename them to "pg<whatever>.dtd", replace the one offending comma with a space, and then validate to that instead.
This is my thought as well. I do think PG-TEI should, at least for its "basic" vocabulary, pick a subset of TEI (subset the elements, the attributes, and constrain/select particular attribute values), and build a flat (non-modular) DTD for that (Pizza Chef will make this almost trivial to do.) And certainly one could tweak the content model (which should be rare) as seen fit (such as tweaking the content model for <div> to include a final <p> after a child <div>.) The resulting DTD won't be a pure subset of TEI P4X, but very close to it. So long as each deviation from pure TEI P4X is well-documented and cogently explained, I don't see anyone having a problem with that. Jon