On Oct 18, 2006, at 5:35 PM, Robert Grimm wrote:
In summary: Yes, there can be subtle ordering problems, but they
usually show quickly during testing.
Is this more of a problem in trying to map existing programming
languages whose native grammars are less rigorously defined onto the
PEG formalism? (i.e. Is there a 'PEG Methodology' that offers set of
heuristics we can use when devising new born-PEG-native grammars to
avoid, or identify and systematically deal with, these issues up
front at grammar design time before we enter testing mode?)
Is there any reported work yet (or projects in the pipeline) on new
PEG programming language grammars based on Bryan's insights from his
dissertation on the new kinds of PL constructs that PEGs could support?
Also, has anyone tried applying PEGs to otherwise potentially
ambiguous NLP phenomenon where a single canonical reading would be
desirable?
This is an area where we might be able to enhance PEGs by leveraging
Steedman's work on CCGs which map complex NL surface structures
directly to logical forms that could be of use in a PL context --- a
topic we hope to take up at The IEUC in future research as soon as we
can find someone with a grounding in both grammar theories who can
make the call on their degree of potential compatibility.
Peter
--
_________________________________________________________________
Peter J. Wasilko, Esq.
J.D., LL.M.
Executive Director, The Institute for End User Computing, Inc.
Visit us on the web at: <http://www.ieuc.org>
_________________________________________________________________
The Institute for End User Computing --- Pursuing Secure, Simple,
Supple, & Sophisticated Systems to Unlock Our Human Potential
_________________________________________________________________
The Institute is a 501(c)(3) tax exempt organization incorporated
under New York State's Not-For-Profit Corporation Law
_______________________________________________
PEG mailing list
[EMAIL PROTECTED]
https://lists.csail.mit.edu/mailman/listinfo/peg
