[Elsnet-list] Boston: AMTA 2006 -- Call for Panels and Roundtables

Priscilla Rasmussen rasmusse at research.rutgers.edu
Fri Apr 14 21:07:59 CEST 2006

AMTA 2006
August 8-12, Cambridge, Mass.

AMTA 2006 is inviting proposals for panels and roundtables to be held
during the main conference days, Wednesday August 9 through Friday
August 11, 2006.  We expect to have up to 3 or 4 such sessions during
the conference.  


Deadline for Proposal Submission: May 7, 2006
Notification of Acceptance: June 8, 2006
Deadline for Final Panel/Roundtable Description: June 30, 2006


Proposal submissions for panels/roundtables should minimally include:

1) Contact information (e-mail and telephone) of the proposer.
2) The topic or theme of the session.
3) The proposed structure of the session (session leadership,
participant presentations, interaction among participants and between
participants and the audience); if applicable, the process for
fostering interaction among participants prior to the conference.
4) Whether prospective participants have been identified and, if so,
their names, affiliations and expected contribution.  

Proposals should be sent to the Panels/Roundtables Chair, Violetta
Cavalli-Sforza (violetta at cs.cmu.edu).  Their receipt will be
immediately acknowledged.  The Chair will then communicate with the
proposers to discuss any issues that still need addressing or aspects
that need further elaboration.  Once proposals have been reviewed by
the Chair and other members of the organizing committee for AMTA 2006,
notifications of acceptance will be sent out, together with
suggestions for revisions.  When the final form of accepted proposals
is submitted, it is expected that the list of participants will be


Both panels and roundtables are intended to present to spectators a
panoply of viewpoints and concerns.  Panels are typically more
structured, with presentations by each panelist followed by questions
and discussions among panelists and spectators.  Roundtables typically
have freer interactions, with specific issues related to the main
topic introduced freely and explored and discussed by roundtable
members and spectators.  Different options for structuring the
interaction in advance and during the panel/roundtable in order to
make the exchange maximally productive, are presented below as


In case you are thinking about proposing a panel or roundtable, but
are not sure whether the subject would be of interest at the
conference, here are a few potential themes/topics that appear to be
of current or ongoing interest.  Other ideas and suggestions are more
than welcome.

1. What are the limits of MT without linguistic knowledge and how do
   we know? What are the pros and cons of different approaches and are
   certain approaches better than others for different applications? 

2. Who are the users and what are the uses of MT systems now, and is
   the state of the art in MT good and cheap enough for them?  What
   applications is MT becoming indispensable for that might accept a
   higher price tag? Are there areas of application of MT that have
   not received sufficient attention, and how can those markets be
   opened up by good enough MT? How do MT research and development
   efforts need to change to support such applications?  Can MT and
   smaller or more specialized tools be financially viable, or are
   they already?

This is a very wide topic in which several subtopics could be
emphasized, possibly in different sessions, for example:
* Different requirements for use of MT: how useful is MT output and
how to identify thresholds for usability in post-editing or
information gathering and other applications.
* The time, cost, and politics of integrating MT into high-volume
production translation.
* The pros and cons of generic MT, i.e. using an MT system the way it
comes from a vendor vs. customizing it for a particular application,
and what customizations might be most useful for different

3. Why are professional translators not using MT even other tools such
   as TMs more?  What are the practical and psychological barriers to
   the use of such tools? What educational structures within and
   outside conventional institutions need to be put in place in order
   to overcome those barriers and make effective use of existing and
   developing technology even if it is far from perfect?  And what MT
   and related tool development and research efforts might allow at
   least some aspects of MT to become more useful to translators? 

4. Can MT be deployed to serve needs of minority or neglected
   languages, and what other data, tools and technologies can be
   harnessed for this purpose? 


An important criterion for evaluating the success of a panel or
roundtable is whether, in addition to presenting multiple perspectives
to the audience, it creates a lively exchange and raises provocative
questions.  In addition, the session at the conference can be more
productive if the participants have started the discussion and
exchanged among themselves ahead of time.  The following is a short
list of ways in which the interaction prior to and during the
conference could be guided and structured for this purpose.  Proposal
submitters should feel free to propose other alternatives, keeping in
mind the goal of achieving maximal exchange among panel or roundtable
members, as well as with the audience.

1. Set up the panel/roundtable as a mock debate or client-customer
   dialogue between the parties (e.g. users
   vs. developers/researchers, developers vs. researchers).  What do
   clients want or don't want, like or don't like, need or don't need?
   Allow some participants to play the role they normally play but ask
   others to take the other side.

2. Prepare a list of questions to distribute to participants ahead of
   time.  Ask them to write a 1-2 page response to those
   questions. Circulate the responses among other participants ahead
   of the conferences to stimulate new questions, new responses and
   material for discussion.  Present the result of the process,
   including the dialectic interaction, at the conference. 

3. Set up an assertion for the panel/roundtable as a topic of debate.
   Each participant must then develop multiple arguments both in favor
   of and against the assertion.  As in 2 above, distribute arguments
   in favor of and against of the assertion to other panelists, in
   order to stimulate debate. Arguments and counterarguments are
   presented at the conference and can be further developed during the

More information about the Elsnet-list mailing list