[logon] HandOn release schedule

[Stephan Oepen]

> I tried this  just now, and ascore wasn't being stored for mcheap
> (perhaps I am too fast :-).

way to fast.  i had not yet built those new run-time binaries.  also
note that the difference between `cheap' and `mcheap' is the precision
on what they report as :score.  [incr tsdb()] now maps that old format
into the newer (more generic) :flags format in the profile, but only in
translation mode.  i guess i should also do that for parsing, but that
will have to be another build then ...

so, when i now (with an up-to-date SVN tree) do

  $LOGONROOT/batch --binary --jaen mrs

then the `result' relation contains valid :ascore values.  truncated to
a single digit, still, but i believe moving to `mcheap' will fix that.
would you agree to me switching the :jacy cpu definition over now?

> I also note that for jacy it is always null ((:ASCORE)), even though I
> load a parsing model.

that is too little information.  i assume while parsing with JACY, but
which parser, and how is it invoked?  i suspect you are looking at the
LKB here?  remember that parse selection in the LKB is only active for
packing mode, with selective unpacking, and running in [incr tsdb()].
effectively, the [incr tsdb()] wrapper around the LKB parser calls the
selective unpacker, which in turn calls *unpacking-scoring-hook*.  the
old `dfki/jacy.lisp' client definition seems to activate all that, but
there currently is no cpu definition to use that for parsing; would you
like me to add a cpu :jacy at lkb for this purpose?

berthold, this is relevant to DeEn too, where currently the LKB parser
is used, but packing (and thus parse selection) are not turned on.  i
quickly tried changing that in `dfki/gg.lisp', but the result is a loss
of coverage on the MRS test suite.  i suspect that means there are some
remaining issues with packing and selective unpacking (in the LKB) with
GG.  are you aware of these?  i would be curious about the reasons.

                                                        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 ---
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++