[
Lists Home |
Date Index |
Thread Index
]
At 05:45 PM 7/30/2003, Gregory Murphy wrote:
>On Wed, 30 Jul 2003, Jonathan Robie wrote:
>
> > What are the relative advantages of these two DTDs? What are the relative
> > advantages of the set of tools available for each? Does either have an
> > advantage in terms of extensibility or compositionality with other
> schemas?
> > Are there other schemas I should be considering?
>
>Each differs in its primary intent: DOCBOOK is for technical manuals, TEI
>for extant books, especially those of scholarly interest. As a result, TEI
>has phrase-level elements for marking up things like variants and errata,
>and has special purpose structural elements for encoding verse and theatre.
>DOCBOOK, on the other hand, has elements for capturing code, screen output,
>commands and their arguments, etc.
*** BOING ***
I knew that, but the significance of it did not hit me like it should have.
I had sort of stupidly assumed that all three of these DTDs were so general
they should all work more or less equally well for general purpose use. But
they each define general purpose use differently. (Q: "What's the optimal
DTD?" A: "What are you trying to optimize?")
Most of general office documents are more like what DOCBOOK describes than
what TEI describes. When I write a memo or an article, I don't want to
think about things like textual variants. I also don't want to think about
most of the stuff in XHTML, I am only interested in the structure.
An interesting benchmark I discovered while working my way through this: I
made a small outline, went down to a place where I could enter phrase-level
text, and counted the number of relevant elements vs. irrelevant elements
for several of the kinds of documents I commonly write. This measures the
signal/noise ration of a DTD for a given purpose. For my own use at work,
Simplified Docbook does very nicely.
Surely someone has come up with a bunch of "rule of thumb" metrics like
this to compare DTDs for a given purpose. Does anyone know where I can find
a good writeup of such metrics?
TEI does nicely for Greek manuscripts, which I sometimes play with in my
free time. But it is not as good for the kinds of articles, memos, and
specifications I spend much of my time writing at work.
Jonathan
|
