[sc34wg3] revised draft Reference Model document N0298

[Bernard Vatant]

> At 10:48 12/04/02 +0200, Bernard Vatant wrote:
> > One remark from Patrice - among other ones - the use of "assertion" might
> > be misleading for people with background in formal logic.
> > Why letting down "association"? What about "statement"?

*Steve Pepper
> Any term we use is going to be "misleading" for someone. (Think of all the
> confusion we will cause among bankers if we use the term "statement" :-)
>
> I'm happy with "assertion". It's already been in use a while, albeit
> informally (e.g. in phrases such as "any time we assign a characteristic to
> a topic we are essentially making an assertion about its subject...").

OK. Then we have to be crystal-clear of our "contextual" definition of "assertion".

> I think it is important *not* to use "association", because from the point
> of view of the SAM, an association is just one kind of assertion (others
> being names and occurrences). We should avoid having a general meaning in
> one model and a specific meaning in another.

Agreed. Well, sort of. In fact I'm used to think about names and occurrences as
associations, no more no less, and never figured why they deserve a specific treat. But if
SAM needs the distinction, clearly we need different terms.

Bernard