The following list shows the correspondence between versions of the IE
task definition and stages of the MUC-7 evaluation.

Version #	Stage

4.1		training and dryrun
4.2		formalrun
5.1		final

The dryrun and formalrun have different domains; the dryrun (and
training) scenario is aircrashes and the formalrun scenario is missile
launches. The final version updates especially the Template Relations
portion of the guidelines.

For both NE and CO, the guidelines remained the same throughout the
evaluation cycle. Their versions are 3.5 and 3.0, respectively.

The format of the guidelines is FrameMaker unless it states that it is
the postscript version. The tar'd files contain multiple files making
up a FrameMaker book. In a few cases, a text version is
provided. Template relations were the last task to be defined and
started with only the BNF being specified.

Normally for each scenario two datasets are provided: training and
test. When the evaluation cycle begins, the label for the scenario
dataset is training then the corresponding test daatset for that same
scenario is used for the dryrun testing. For the formal run, a formal
training set is given out 4 weeks before the test answers are due
in. The formal test is given out one week before the test answers are
due. After the entire evaluation and meeting have been held, final
edits are made if necessary.

The tokenization rules and the time supplement are considered
supplementary only because they deal with smaller details. The rules
in these documents were followed during the course of the evaluation.