On Sun, 24 Nov 2002, Adrian Midgley wrote: ... > > One difference - Gary reviewed specific existing systems for their > > utility. He did not just propose to create a laundry list of requirements. > > Gary has a nice position in that the requirements in theatre and > peri-operatively are more clearly specified than in wider areas.
Hi Adrian, I feel that it is more accurate to say that Gary put himself in a nice position by clearly specifying a problem area :-). In fact, there are numerous examples from various scientific disciplines that support the importance of defining the problem well. > Part of the lesson from the Exeter Project in about 1970-80 was that > different specialties need rather different main views. This is indeed a very valuable lesson. As I mentioned during the OIO "future-proof" talk, I chose to define OIO's target problem as supporting diverse needs and changes-over-time. A "common compromise for expediency", in my opinion, is a set of pre-defined "main views" and lack of support for the creation of additional and alternative "views" and data fields. I believe GnuMed does try to address this - as does the OIO system in a big way. However, TkFP, OSCAR, and even VistA are also extensible (in data and view). In particular, VistA has a modular architecture with a formal naming convention for locally added/forked modules. Maybe the VistA experts / Hardhats will tell us more about these features? ... Best regards, Andrew --- Andrew P. Ho, M.D. OIO: Open Infrastructure for Outcomes www.TxOutcome.Org
