No, good point.

Thanks,

Matt
Dave Reynolds wrote:
> Matt Williams wrote:
>
>
>> Ummm...maybe. I would interpret all of these ( ->) as being defeasible
>> inference - in which case 'says/ claims' would seem to be a reasonable
>> interpretation of the arrow (so A suggests/claims that B suggests C).
>>
>
> OK. I can see that you want to capture situations where the thing claimed
> is also defeasible - either because it's a claim about what some other
> paper claims or because it it is an inference which is only true if some
> unstated other things occur (e.g. Newtonian physics as an approximation
> only valid at low speeds).
>
> It just seems to me that there may also be times when the paper is claiming
> a hard inference:
>
> A claims (B -> C)
>
> and then if you find a situation in which B is true and C is false then you
> know it is A's claim that is refuted.
>
>
>> This is based on argumentation (one of the main topics of my PhD, I
>> hope). I've seen Claimaker, although I've looked more at Araucaria
>> (http://araucaria.computing.dundee.ac.uk/). The big problem (for me) is
>> that although they look lovely, there isn't a 'handle' to crank. OTOH,
>> the implemented argumentation stuff is very sparse (esp. WRT ontological
>> info.). So I'm trying to implement something that isn't as rich as
>> ClaiMaker, but has a little more going on behind the scenes.
>>
>
> Sounds good fun, hope it goes well.
>
> Dave
>
>
>
>
> Yahoo! Groups Links
>
>
>
>
>
>
>
>