[Attempto] Draft paper for discussion

[Norbert E. Fuchs]

On 26 Apr 2010, at 16:16, Adam Wyner wrote:

> ...
> http://wyner.info/research/Papers/WynerVanEngersBahreini2010.pdf
> 
> Adam Wyner, Tom van Engers, and Kiavash Bahreini
> Submitted to eGOVIS 2010

Adam

I had a first look at your paper and would like to add two comments to your statement

> However, as a first-order reasoner, RACE cannot reason with modal operators such as should or verbs which take sentential complements such as say; to reason with these expressions requires a modal logic, which has not been implemented with ACE.

ACE provides modality with modal auxiliaries for possibility (can/cannot/can not/can't/it is possible that/ it is not possible that), necessity (must/have to/does not have to/ it is necessary that/ it is not necessary that), recommendation (should/should not/shouldn't/it is recommended that/it is not recommended that), and admissibility (may/may not/it is admissible that/it is not admissible that). Furthermore, transitive verbs can take a complete sentence as an object.

The latest version of RACE – not yet publicly available – provides reasoning with modality (possibility, necessity) and with sentential complements. This reasoning is implemented via the standard translation of modal logic into first-order logic, i.e. using possible worlds semantics. The modal auxiliaries for recommendation and admissibility are not covered by RACE since they do not seem to have an accepted translation into first-order logic.

I'll will read your paper more carefully in the next days.

Regards.

   --- nef