The specialization feature of DITA allows for the creatio
of new element types and attributes that are explicitly and formally
derived from existing types in a way that allows for the blind interchange
of all conforming DITA content and a minimum level of
common processing for all DITA content.
Specialization DTD can be used to create a base for a structural
type spcialization with the correct naming and modularization conventions.
This base can then manually edited to define the element level
A shell DTD defines the set of structural modules
(topic or map types), domain modules, and constraints modules that the map
or topic can use. In general, any production use of DITA involves
definition of the set of modules best suited to local requirements. This
always requires the creation of local shell document types, even if all
they do is omit unneeded modules or apply constraints to the standard
about the benefits of creating local shells.
Domain attribute specialization allows DITA document type
developers to incorporate new conditional processing attributes that can
be used for filtering and flagging, or new attributes with no existing
equivalent that can be managed and generalized in the same way as
conditional processing attributes.