|
The
Annotated Text Processor (ATP)
| Back |
ATP Model
for IDD Dictionaries |
Next |
(click
to enlarge)
ATP was designed to be an interlinear
text processor but ATP's data-modeling capabilities
support a much wider range of document types and applications.
In principle this means that ATP writes interlinear
paragraphs and title and author paragraphs. We are envisioning
modifications to ATP's paragraph-formatting algorithms
that should allow us to support many other kinds of
paragraphs and thereby support many other kinds of documents.
In the near future, we plan to add plain-text paragraphs,
tables, lists, and hierarchical-entry paragraphs characteristic
of dictionaries and catalogues, in which many distinct
data types are entered recursively and concatenated
inline.
This example shows a model for an IDD
dictionary which is conceived as a wordlist or entry
paragraphs. This prototype ATP model for IDD dictionaries
will load contents for a given entry from all the relational
tables of IDD into nested repeating elements within
a single parent entry. English-to-original-language
wordlists will be constructed out of those same entries
using information in appropriate elements (englishkeyword
etc.) within those entries.
Proposed revisions to ATP include appropriate
support for sorting document contents and maintaining
and exploiting internal links and cross-references.
The proposed revisions should make
it possible to create active research and working environments
for projects like grammars or historical semantics.
To consider such possibilities we have included an prototype
model for a grammar in the ATP tour.
Back
to Annotated Text Processor
|
