On Expr and Fun, @per="effect" is not allowed by original XSD about reaction-ruleml HOT 10 CLOSED

The value "effect" for the attribute @per is not used on the DR layer of Reaction RuleML, because there are no actions and functions with side-effects (e.g. procedural attachments) on this layer of pure reasoning dialects. Side effects are introduced later with actions and events in the active dialects. It then uses the per-effect.attrib group instead of the per.attrib group which is defined in the attribute_module in the XSD schema.

from reaction-ruleml.

IC. I need to take "effect" away from Rels then also (in the RNC).

from reaction-ruleml.

I believe the XSD also has @per="effect" allowed for Rels also - I'm pretty sure that is why I put it in the RNC in the first place.

from reaction-ruleml.

Adrian- the other new value for @per is "modal". Do you have an example
of that usage?
Since modal operators take formulas, it is not clear to me how a modal
operator is supposed to be applied (syntactically).

Tara

On 3/28/14 9:28 AM, Adrian Paschke wrote:

The value "effect" for the attribute @per https://github.com/per is
not used on the DR layer of Reaction RuleML, because there are not
actions and side-effect full functions on this layer of pure reasoning
dialects. Side effects are introduced later with actions in active
dialects. It then uses the per-effect.attrib group instead of the
per.attrib group which is defined in the attribute_module in the XSD
schema.

—
Reply to this email directly or view it on GitHub
#43 (comment).

from reaction-ruleml.

I now remember why we need effects also in the DR dialect. Although there are no explicit actions it supports object constructors which have an effect - http://ruleml.org/reaction/0.2/examples/procedural_attachments.rrml.

That means @per="effect" needs to be supported also in expressions. I need to correct this in the XSD. And it should be also supported in the RNC.

from reaction-ruleml.

The RNC currenty supports @per="effect" in Expr, Fun and Rel - no change is necessary to the RNC regarding this issue.

from reaction-ruleml.

I have applied the upgrader 0.91 to 1.0 (http://ruleml.org/1.0/xslt/upgrader/091-to-100.xslt) to the procedural_attachments.rrml file, and added it to the exa directory.

from reaction-ruleml.

With commit d35cf5a , this is resolved.

from reaction-ruleml.

This latest commit for adding @per="effect" to Fun and Expr in the XSD also allows @per="modal" to Fun and Expr, since the same attribute group is used for Fun and Expr that is used for Rel. Is this the intended result?

from reaction-ruleml.

I added an example "ModalDerivationRule" for the modal interpretation in the exa folder using a generic connective operator as a modal connective. Formulas are then under modal interpretation (@per="modal").

In the KR dialect modalities can be represented as reified fluents using expression functions with @per="modal". This is used e.g. in the RBSLA ContractLog KR (http://rbsla.ruleml.org) for Event Calculus based deontic reasoning.

from reaction-ruleml.