Sandro Hawke wrote: >>The negative entailment tests are reported as passing in the weak sense that >>we >>do not find the entailment. These should probably be reported as incomplete >>rather than pass. > > They should indeed; will you please change this? Done. > As I understand it, passing a N.E.T. means proving that the entailment > does not hold in the given logic (DL, Full, Lite). Calling it > incomplete at least says it's not a "Fail", and the output link can > give details about how thorough the search for a decision was (even > though finding one would have indicated a failure). Or you can just > skip them, since you know it'll never pass. I've chosen to skip them since that seems to be what other similar systems have done. DaveReceived on Wednesday, 24 September 2003 17:19:45 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:19:15 GMT