Hi Lee, Am 11.10.2011 um 18:41 schrieb Lee Passey: [snip cite from Marcello]
Now Marcello, there is no need to resort to ad hominem. The fact of the matter is that even if BowerBird had consistently behaved in an exemplary and respectful fashion, z.m.l. is still inadequate as a markup language.
Part of the problem is that we have no real specification of what z.m.l. /is/. BB has provided tons of "examples," (some of which are now inconsistent as apparently the language is still evolving) but no definitive declaration of what the language allows, and disallows, and how the elements are to be used. This is why I have dubbed z.m.l., SML -- Spousal Markup Language: there are rules, but you have to figure them out on your own, and they are subject to change on a whim. I have read the specs to TEI, I have, also, read how text is supposed to be marked up with TEI.
Talk about "Spousal Markup" and inconsistancies. TEI is far to bloated to be useful.
The original PG philosophy was that the text was the only thing that mattered, and all markup was superfluous. It quickly became apparent that at least emphasis needed to be indicated and so it was decided that italicized text would be indicated in UPPER CASE. Unfortunately, people began to discover that there were books which contained upper case text which was not intended as emphasis, so the /new/ standard became to use underscores to indicate italicization (only those of us old enough to have learned to type on typewriters will recall that the mechanical convention of typing was to underline what would otherwise be italicized).
There are a number of constructs in other markup languages which z.m.l. does not support. BowerBird's response is that support for those constructs is unnecessary as e-books simply do not require them. This is, of course, the same argument as the one that /all/ markup is unnecessary, the line is simply drawn in a different place, and BowerBird becomes the ultimate arbiter of what is, and is not, needed in e-books.
It's hard to know what markup will be necessary to preserve any specific work of literature. Thus, what is really needed is an eXtensible Markup Language, such as TEI, which captures everything we know about now, and can be extended when we encounter something new. z.m.l. fails on both these counts.
I do agree that some mark up is required and a decent spec is needed. But, there is a problem with a extendible language. They more than often allow the user to have several avenues to extend the language of any particular feature. This then causes inconstancies with the usage of the mark-up and different users do not have a consistent way of marking up said features. I would propose an modular language that is extended when the need arises with only one way to represent a particular feature. The modular approach has the advantage that the tool chain can be modular in design and the tool chain need not be rewritten to support the new "feature", but just needs a new modul to handle it. Furthermore, the master format should preserve the original structure of the text while not being so restrictive to not allow the restructuring of the text. That is these mark up elements can be ignored or interpreted differently in order to support another output format. regards Keith.