I see two distinct "things" in OSLC Automation that will be related to the section of the "Actions" spec that will (/may) be in Core: 1. The AutomationPlan implementation type of Actions. This defines how AutomationPlans can be used as implementations of Actions, and this definition can be re-used by any other domain or spec. 2. The Automation profile of Actions. This is how Actions are to be used in the Automation domain (and domains that extend it). This includes the restriction that all executable Actions on Automation resources MUST have an AutomationPlan implementation. It is unlikely that other domains/specs that use the AutomationPlan implementation type will also use these restrictions, which is why I think it's important to distinguish the two, and not include the definition of that implementation type in what we call the "Automation profile [of Actions]".
Also, based on John's recent emails, I want a new term for what I've been calling "non-parameterised plans/actions" when referring to actions in the actionOnResult property on AutoPlans. In place of that term, I am now considering saying "future-Action" or "contextless Action" (an action that will later have a context resource, but does not have one yet). (This may suggest that we introduce a specific rdf:type value for that resource if we now give it a distinct name). I will now use the terms "parameterised plan/action" and "non-parameterised plan/action" to solely refer to plans that have/don't have parameters _in addition to_ their context resource. I'll talk about this in the call shortly, but I wanted to send it out to the list too. Martin Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU
