Andrew Ho wrote:
>>Otherwise, tell me the difference!
>>
>One difference is that GEHR has decided to use Eiffel. The GEHR team
>
this is just our technology choice. There is no dependance on Eiffel of
GEHR.
>questions whether GEHR should be implemented with Zope/Python, Java, etc.
>
I question the quality of Java, but only in the same way as I would for
any other software - i.e. not GEHR-specific. Zope and Python I know
almost nothing about.
>OIO is currently implemented with Zope/Python/SQL. I am pretty sure that
>OIO can also be implemented with Java, Eiffel, C, Perl, etc.
>
I don't think you need to change, if what you have works! Our main
concern should be data quality. As long as there is never a danger of
creating a term "kidney, left" instead of a "kidney, right", or adding a
0 to a chemotherapy prescription, or mixing the percentages on a
differential diagnosis or whatever, we are basically ok.
>GEHR/OpenEHR is perceived to be "great technology but a few years away".
>
We'll have an openEHR kernel running sometime this year.
>OIO has been in use for 2 years and *I think* is functionally quite close
>to a GEHR kernel. We even have an online metadata repository running for
>over a year now (OIO Library).
>
which is great experience, and when we have the resources, we need to
study the ever-evolving OIO system.
- thomas beale
