Two reasons we are optimistic about using literate programming for
this application:
first, linguists (unlike computer programmers) have a more than two
thousand year history of writing grammars for people to read, so it's
not hard to get them to write prose; and second, grammars
(particularly the formal grammars) tend to be shorter than most
computer programs these days, so the idea of doing literate
programming for grammars is more practical than it would be for, say,
the Linux OS.
The citations (with URLs for the PDFs):
Maxwell, Michael, and Anne David. 2008. Interoperable Grammars. First
International Conference on Global Interoperability for Language
Resources (ICGL 2008), Hong Kong. http://hdl.handle.net/1903/7568.
Maxwell, Michael, and Anne David. 2008. Joint Grammar Development by
Linguists and Computer Scientists. Workshop on NLP for Less Privileged
Languages, Third International Joint Conference on Natural Language
Processing, Hyderabad, India. http://hdl.handle.net/1903/7567.
Some older ramblings on this idea are the following (but some of the
ideas have been superceded, in particular the idea of using the xfst
programming language for the formal grammar):
Amith, Jonathan D., and Michael Maxwell. 2005. Language Documentation:
The Nahuatl Grammar. In Alexander Gelbuck (ed.) Computational
Linguistics and Intelligent Text Processing. Lecture Notes in Computer
Science. 474-485. Berlin: Springer. http://www.springerlink.com/content/26tpwwnptltcvjy8/.
Mike Maxwell
CASL/ U MD