This may be part of the deferred namespace discussion (since I think
that namespaces would be required to address my question), but the DITA
spec says the following under "Class attribute syntax":
"Each value has two parts: the first part identifies a module package,
for example a topic type or domain package name..."
The discussion defines "module file" (which I think should be "module
declaration set" to get away from use of the term "file") but it doesn't
define "package" or "module package".
In addition, I'm not seeing anything about how the module package name
used in the class attribute relates to the storage objects that make up
the module's declaration set--in most cases in the DITA-defined
materials the module name is the same as the module's top-level
declaration set's file name which is also the name of the topic type
being declared (for topic modules--obviously that won't work for
domains), but not always (glossentry in glossary.dtd).
It feels like there needs to be some way to declare module names but
where? For example, within XSD schemas we could define specific markup
to put within annotation/appinfo for formally declaring module names,
but there's not much we can do for DTD declarations beyond PIs or
conventional comments. [And we've already discussed the need to be able
to namespace quality both module names and type names w/in class, but
we've deferred that discussion for later.]
The spec should probably say something to the effect that module names
are not constrained just so it's clear that one can choose any name for
your module.
Cheers,
E.
--
W. Eliot Kimber
Professional Services
Innodata Isogen
8500 N. Mopac, Suite 402
Austin, TX 78759
(214) 954-5198
ekimber@innodata-isogen.com
www.innodata-isogen.com