Determinable properties

[Acilio Mendes]

There is a native way to express determinable properties on NARS,
without resorting to the relations?
For example, how can I express:
"A object has a attribute 'colour', so that this colour can be red,
blue or green." or
<(&&, <{#} --> object>, <(*,{#},colour) --> hasAttribute>) ==>
(||,<(*,{#},colour, [red]) --> hasValue>,<(*,{#},colour, [blue]) -->
hasValue>,<(*,{#},colour, [red]) --> hasValue>)>.
"Humans have a attribute 'name', wich is composed by a first name and
family name."
without depending on the experience-grounded meaning of the systems?

--
[]´s
Acilio.

[Pei Wang]

On Wed, Oct 20, 2010 at 7:29 PM, Acilio Mendes <acilio...@gmail.com> wrote:
> There is a native way to express determinable properties on NARS,
> without resorting to the relations?
> For example, how can I express:
> "A object has a attribute 'colour',  so that this colour can be red,
> blue or green." or
> <(&&, <{#} --> object>, <(*,{#},colour) --> hasAttribute>) ==>
> (||,<(*,{#},colour, [red]) --> hasValue>,<(*,{#},colour, [blue]) -->
> hasValue>,<(*,{#},colour, [red]) --> hasValue>)>.

The simplest way is probably by introducing ad hoc concepts. Since the
"object" is a special type, it may deserve a dedicated term.

"Attribute" is an artificial notion, which is often unnecessary.

How about the following:

object-s --> object
(object-s * colour-s) --> has-colour
colour-s <-> {red, green, blue}

> "Humans have a attribute 'name', wich is composed by a first name and
> family name."
> without depending on the experience-grounded meaning of the systems?

(human * name) --> attribute
name <-> (first-name * family-name)

The meaning of all terms are experience-grounded. The above statements
can be added into the system as part of its experience. Of course, for
practical purpose, they can be loaded into the memory.

Pei