On 25/05/2010 01:44, Heath Frankel wrote: > > Thomas, > > With respect to having some nodes with an at-code but nothing in the > ontology, I would need to see some rules that can actually work for this. > > */[HKF: ]/* > > To me the rules when a node ID is not required in the ontology are > pretty simple: > > * a node is not of type LOCATABLE >
at the moment, this is always true anyway. In the future, we will most likely need to make PARTICIPATION archetyped, even though it is not LOCATABLE. So I will keep this in mind in the specification and reference tool. > * a node is the root ITEM of an ITEM_STRUCTURE (e.g. items CLUSTER or > item ELEMENT) > > The problem with not having a ontology item for the root ITEM of the > ITEM_STRUCTURE is that we have nothing to use as the default name of > that ITEM. This demonstrates how this level in the RM is semantically > redundant. > Heath, I am not sure of what the issue is here - can you clarify? > We also need to be careful about your removal of node IDs on single > attributes such as description. In a template, a description maybe > filled with an ITEM_STRUCTURE archetype. Therefore in an operational > template, this description node must retain the archetype ID of the > filler ITEM_STRUCTURE. > that is true. Note that I have not 'removed' anything; the ADL workbench just treats these single-attribute nodes with only a single constraint in the same manner as XML - there is nothing to distinguish, so it doesn't waste an at-code, ontology entry, or space in paths unnecessarily. But as soon as their are either multiple alternatives, or it is a multiply-valued attribute, it does require an at-code. But of course you are right about the archetype id needing to be retained in all circumstances, so I will also specifically indicate this in the specification and take care of it in the tool. > Another case for optionally maintaining a node ID of single attributes > is where it is desired to name this node. It is common in templates > to rename the description to something like Medication description > which results in an XML element with this name in a template data schema. > In an ADL 1.5 template, this would cause a new at-code to come into existence at the specialisation depth of the template in question, e.g. it will be a code like at0.0.1. So this would effectively be a redefinition of a non-coded node into a coded one, which is a slightly special circumstance which I also need to take care of in the specification and tool. These are all good points, and just go to show how important implementation is in telling us how to move forward. Would it be unkind of me to say that paper standards developed by committees don't generally manage too well on this score?! - thomas beale * * -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.openehr.org/mailman/private/openehr-technical_lists.openehr.org/attachments/20100525/58ee8b23/attachment.html>
