>> I am having a hard time understanding when the modeller would not know if there was a series of events or not <<
Well for an experienced modeller it is not too hard, although there are still quite a number of grey areas. Heather and I still have discussions and disagreements about the correct Class or use of class attributes. The point is that for beginners or casual users, it *is* hard. Just look at Pablo's or Saso's experience. I want to improve the architecture and, I agree, definitely the tooling, so that people are not forced to think about the hard stuff until they are good and ready or the use-cases emerge, but confident that they are not making a 'wrong' decision which stops their work being shared with a wider community in due course. Pablo's experience is not at all unusual - most groups find it hard to think beyond the boundaries of their own experience, and it is costly to think ahead in terms of design decisions. You forget just how much expertise and detailed consideration that you have given to this problem space. Even with training and experience it will take some time for others to catch up (including myself!). I want people to be able to hit the ground running and not be intimidated by the technology or the ontological philosophy. openEHR does a pretty good job on both counts but I think we can do better. Tooling improvements certainly but tooling is ultimately driven by the RM and based on the experiences of implementation and training over several years I think now is good opportunity to take stock and find some ways of further reducing buy-in without compromising on the excellent work done to date. Ian On 27 March 2012 11:51, Thomas Beale <thomas.beale at oceaninformatics.com>wrote: > On 27/03/2012 10:42, Ian McNicoll wrote: > > Hi Thomas, > > I definitely agree here. While I think there is huge merit in having > some kind of simplified single Event OBSERVATION, there is absolutely a > need to handle the increasing numbers of device mediated multiple event > observations. > > As a modeller though, I really do not want to have to commit to one > class or another. I want to be able to start with a simple model and > 'unconstrain' that when and if the individual or generic use cases emerge. > Given the complexity of multiple event handling I suspect that is a pretty > tough ask but if you don't ask you don't get :-) > > > I am having a hard time understanding when the modeller would not know if > there was a series of events or not. Surely all vital signs should be > modelled as that (even if in some instances in the data, there is only one > event)? There seems no other obvious way to model Apgar or OGTT; I would > expect some ECG and blood glucose to be in time-series form as well. > > That is why I was interested in what Marand are doing which > is essentially flattening or hiding the complexity of the mutliple event > Observation - simple API vs. Complex API where the simple API just hides > the complexity for the 80% of use-cases. Is that feasible? > > > that was actually the intention of the original design - to support APIs > that would allow a 'nice' way to create a single event Observation (indeed, > it should have been defined in the Observation class itself). But if we > want a single-event Observation class as a new variant, no reason why not. > > > openEHR archetypes are still easily the best way to manage the very > difficult task of identifying clinical content requirements and gradually > developing these as new use-cases and complexities appear but I think we > can do more to make that process one where we can start simple and > gradually add complexity in a backward compatible manner. > > > personally I think a lot more work is required on the tooling... > > - thomas > > > * > * > > > _______________________________________________ > openEHR-technical mailing list > openEHR-technical at lists.openehr.org > > http://lists.openehr.org/mailman/listinfo/openehr-technical_lists.openehr.org > -- Dr Ian McNicoll office +44 (0)1536 414 994 fax +44 (0)1536 516317 mobile +44 (0)775 209 7859 skype ianmcnicoll ian.mcnicoll at oceaninformatics.com *Primary Health Info 23 ? 25th April in Warwick ? are you coming?<http://www.primaryhealthinfo.org/> * Clinical Modelling Consultant, Ocean Informatics, UK Director openEHR Foundation www.openehr.org/knowledge Honorary Senior Research Associate, CHIME, UCL SCIMP Working Group, NHS Scotland BCS Primary Health Care www.phcsg.org -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.openehr.org/pipermail/openehr-technical_lists.openehr.org/attachments/20120327/9e89a1a2/attachment.html>
