[Odrl-version2] Proposal: Multiple Actions

[Steven Rowat]

On 8/3/10 8:45 AM, Daniel Pähler wrote:

> ...as far as I can tell, this could only be modelled by
> creating two subclasses of Permission. Till now though, the Core Model works
> without the "subclass-of" relation.

Understood. And I agree, this extra complexity in the core model would 
not be a good thing on its own and would have to be weighed carefully 
against any advantage in otherwise 'simplifying' the code.


> If I understand you correctly, you're suggesting to have Constraint as a
> subelement of Action, as in
>
> <action name="http://odrl.net/2.0/action/display">
> 	<constraint ...>
> </action>

No, my syntax was poor, but I meant just that the constraint would 
follow directly after the Action that it was constraining, as in:

<action name="http://odrl.net/2.0/action/print"/>
<constraint value=''/>

Thus the Constraint would be a child of the Permission, not the 
action; but it would constrain the action immediately before it.

In other words, like in some CSS relations as I recall, the constraint 
would automatically be read as applying to the immediately preceeding 
action. This would require no extra code linking; if there's a 
constraint for an action, it appears immediately afterwards; if 
there's not, it doesn't appear.

Or if it's required to refer to the Constraint for other purposes, it 
might be possible to label each constraint, say by using the final 
term that defines that preceeding action as the identifer; so in the 
example above it might be:

<action name="http://odrl.net/2.0/action/print"/>
<constraint id='print' value=''/>


However, all this may be moot in comparison with your next point, 
which appears to be at a deeper level:

> My intuitive understanding is that only the Permission is constrained, not the
> Action that it allows.

And my intuitive understanding had been the opposite; that it's the 
Action that's constrained (can there be something that is constrained 
without being previously defined as an Action with a Permission?)

But perhaps none of this matters, because after looking at the 2 Core 
Model and XLM schema and seeing Extended Relations (which I had 
forgotten about), I realize I'm in well over my head here since I have 
no idea how to integrate the (proposed?) Extended Relations with 
Renato's proposed compaction into multiple Actions. Would there be 
conflicts? If so, then I'd tend to agree with you that it doesn't seem 
worth pursuing the compaction even for Actions, much less Constraints.

Steven Rowat