[mrp-users] updates from the MRP 2019 shared task

[Stephan Oepen]

dear colleagues,

thank you for your interest in the CoNLL 2019 Shared Task on
Cross-Framework Meaning Representation Parsing (MRP 2019)!  as of
today, this mailing list has some 80 subscribers, and we of course
hope that most of you will end up forming teams and participate in the
task.

we have been updating the task web site with additional information
over the past week, including a semi-formal description of the uniform
graph serialization we will adopt for the task and a publicly
accessible sample of 89 venerable WSJ sentences for which semantic
graphs are annotated in all five frameworks represented in MRP 2019.

for links to these resources, please see the task schedule on the main
task page:

  http://mrp.nlpl.eu

on monday this coming week (april 15), the first release of the task
training data will become accessible for download from the LDC.  to
obtain the data package, each (prospective) team will need to exectute
a no-cost evaluation license: please fill in, sign, and submit to the
LDC the following form:

  http://svn.nlpl.eu/mrp/2019/public/license.pdf

for the time being, we will look at the subscribers to this mailing
list as our running tally of candidate participants in the task.
thus, in case you know of colleagues who might be interested but are
not yet subscribed to our list, please encourage them to follow the
instructions for prospective participants on the task web site.

we look forward to working with you!  for questions or suggestions,
please do not hesitate to contact the team of task co-organizers at
the email address in the Cc: on this message.

best wishes,

omri abend, jan hajič, daniel hershcovich,
marco kuhlmann, stephan oepen, and nianwen xue