Greg Weinger wrote:

> Hello,
>
> I have a basic question concerning the overlap of ADL and OWL.  Since 
> you're looking to export ADL to OWL, it seems like they overlap quite 
> a bit, but what are the differences?

Hi Greg,

I have documented some of the differences in the latest ADL draft - go 
to the drafts page under 'development' on openEHR.org. I am not an OWL 
expert so the University of Manchester people or Mayo group may want to 
add corrections.

In practical terms what we know so far is that:
- OWL does only a few of the leaf constraint types that ADL does, e.g. 
constraints on dates, integers, reals, strings and so on. But we are 
told that this will improve as time goes on
- it is still an open question as to how to represent the ontology part 
of an archetype in OWL; experts in this area are still wrking on it

My gut feeling is that converting an ADL archetype to OWL will 
eventually be doable, but it will be 6-12 months before we get it right, 
since some of the open questions are the subject of research, including 
at Manchester and Mayo. We're really on the bleeding edge here.

>
> Also, from the web site I couldn't tell if the HL7 Structured 
> documents were to be published in *both* ADL and OWL, or in a 
> *combination* of the two.
>
> Please direct me to a URL if I've missed some pertinent documentation 
> (admittedly, I've only read the OpenEHR web site so far, not the 
> seminal Archetypes paper).

In addition to the original paper, I recommend the latest ADL and AOM 
drafts on the development/drafts page - these have significant 
improvements to the current official versions.

We are all hurriedly learning as fast as we can, and the Mayo Clinic's 
project will be a real test bed for archetypes in an OWL environment. As 
Peter has said, there will be software available soon (indeed the 
current ADL workbench and Clinical archeytpe editor contain a rough OWL 
outputter - see the downloads from www.oceanInformatics.biz), but I want 
stress that the output is _experimental_ and curently technically not 
correct  - it will be improved as we go. If you have expertise in this 
area, it would be valued.

Hope this helps.

- thomas



-
If you have any questions about using this list,
please send a message to d.lloyd at openehr.org

Reply via email to