Bowerbird@aol.com wrote:

...

In z.m.l., anything surrounded by two or more blank lines is structured text.

wait a minute! _i_ make the rules for z.m.l., not you. :+)

Sure, but you don't get to call a lemon purple, just because that's the way you see it. So if you require that paragraphs in your markup language be segregated from the surrounding text by one, and only one, blank line you have forced a structure on the text. Hence, it is, by definition, a structured text. [snip]

in z.m.l., blank lines are the very thing that _define_ paragraphs -- and titles too. so i am afraid that z.m.l. will never "be able to" do what you are asking. but...

Precisely my point. I was hoping that you were coming up with an algorithm that would identify _real_ paragraphs (as defined in the English language) inside unstructured text (human beings are pretty good at it, but even they are not infallible), and instead you have told me that you can detect blocks of random text which are delimited by blank lines (which we shall refer to as bns-paragraphs, for "bowerbird new-speak paragraphs") from inside of z.m.l.-structured text. The µBook ebook reader can do that right now (PG simplified text looks really good in µBook). I can't see why this particular wheel needs re-inventing.