[developers] SOLVED: Re: Mysterious generation problem in LOGON

[Berthold Crysmann]

Dear all, 

today I discovered the reason for the strange behaviour I reported a few
days ago. 

1. The reason for the difference in behaviour between the source and
binary version of LOGON was a configuration issue on my side. 
:logon was not pushed as a feature, and a look at the code revealed that
score-strings indeed behaves differently depending on this feature. 

Once recompiled, score-strings worked in the source version, and
generation of wh-matrix questions seized to work. 


2. The difference between the LOGON and the standard version of LKB
appears to be real: when confronted with an unspecified PRED value (in
C-CONTS), standard LKB still happily generates (the value got specified
by some other rule) as expected, whereas LOGON does not. I suspect this
is due to a difference in indexing, but - not being an LKB developer - I
can only speculate here. 
Introducing a more specific PRED value in the wh-h rule solved the
problem for LOGON as well. 

PS: Could there not be a warning message, whenever PRED predsort is
detected?

Cheers, 

Berthold


On Tue, 2008-01-08 at 21:12 +0100, Berthold Crysmann wrote:
> On Tue, 2008-01-08 at 09:24 -0800, Stephan Oepen wrote:
> > hi berthold,
> > 
> > > cd $LOGONROOT
> > > bin/logon --source
> > > 
> > > (load "lingo/lkb/src/general/loadup")
> > > (load "dot.clinit.cl")
> > > (compile-system "fad")
> > > (compile-system "mt")
> > > (compile-system "tsdb") 
> > > 
> > > 
> > > Unfortunately, the function mt::score-strings does not work as expected,
> > > then. 
> > > 
> > > The sentences that mysteriously do not generate include all wh
> > > questions...
> > 
> > so, what exactly happens?  any calls to mt::score-strings() fail?  or
> > just some?  give me an actual example to debug!
> > 
> 
> If I invoke logon as $LOGONROOT/bin/logon without the --source
> parameter, i.e. the runtime, I only ever get (0) as the result of
> score-strings. If, however, I use the runtime, I get far more
> interesting values. Unfortunately, though, I lose readings in
> generation, then. Inspecting the generator chart, the crucial rule wh-h
> is missing, although I can manually unify it. I checked for the usual
> suspects like index filtering, but to no avail. 
> 
> BTW, the version of ACL we have at DFKI is 8.0, in case that matters... 
> 
> Do you need a version of the grammar?
> 
> I can send you a .tar.gz, if you want.
> 
> Berthold
> 
>  
> >                                                        cheers  -  oe
> > 
> > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> > +++ Universitetet i Oslo (IFI); Boks 1080 Blindern; 0316 Oslo; (+47) 2284 0125
> > +++     CSLI Stanford; Ventura Hall; Stanford, CA 94305; (+1 650) 723 0515
> > +++       --- oe at ifi.uio.no; oe at csli.stanford.edu; stephan at oepen.net ---
> > +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
-- 
PD Dr. Berthold Crysmann
Sprache und Kommunikation
Institut für Kommunikationswissenschaften, Universität Bonn
Poppelsdorfer Allee 47, D-53115 Bonn