Hi Dave,
Can you point me to a description of what constitutes a "standard" interface for CORBAmed? Is there a detailed "HowTo" somewhere? Also, do you know of an implementation on Palm Pilot?
The "standard" interfaces are in the CORBAMed interfaces (IDL). You can look at them in the src/interfaces/corbamed/idl directory (of OpenEMed). They get translated from IDL to Java by the "idl to java" compiler.
That code is in src/interfaces/corbamed/src directory. Documentation on all the interfaces is in our docs directory.
If your Palm Pilot has Java, enough memory and a network connection, you can run OpenEMed on it!
I imagine it would be quite nice to be able to have OIO on Palm Pilot and be able to "sync" to a remotely located server through CORBA. Has that been done before? Is it possible?
This is entirely possible. The only constraints are memory and network connectivity.
Dave
Thanks,
Andrew
---
Andrew P. Ho, M.D.
OIO: Open Infrastructure for Outcomes
www.TxOutcome.Org
Assistant Clinical Professor
Department of Psychiatry, Harbor-UCLA Medical Center
University of California, Los Angeles
On Mon, 19 Feb 2001 21:36:43 David Forslund wrote:
>At 03:55 PM 2/19/2001 -0500, John S. Gage wrote:
>
>>>Good idea. There is no reason for OIO to re-implement CORBAmed. So far,
>>>we have not encountered an application that requires CORBAmed.
>>
>>I believe, and David Forslund will correct me, that CORBAmed is intended
>>to solve the perceived interoperability problem in medical computing.
>
>This is basically correct. A simple way to think of this is to reduce the N
>X M problem to a N X 1 problem. It isn't clear that OIO has faced the
>problem of having multiple systems talk together. That is when the OMG
>standard interface
>approach becomes most useful.
>
>
>>>In addition, I am not sure how best to incorporate code from OpenEMed
>>>into OIO, yet. Do you have any suggestions?
>>
>>Hmmmm. I now wonder who(m) is being Socratic. Of course, I have a
>>suggestion: implement everything in Java. If you implement OIO for the
>>Palm in Java, then use that code to replace the Zope.
>
>You don't need to implement OIO in Java, just implement the various
>standard interfaces in Python (which is the language of Zope, as I
>understand), and you are ready to go.
>
>>One of the issues that I am interested in is what relational database are
>>you using, or do you store data in some other format?
>
>The point of the COAS interface is to avoid having to know what
>relational (or other) database anyone is using.
>If a third party has to know what database you are using, then we have not
>solved the interoperability problem, because they will have to figure out
>how to map to their database (N X M). If you map to the standard
>representation (including methods), there is a lot less work. Not really
>any different for the first translation, but that should be all that you
>have to do rather than having to do it for everyone that comes
>along. Other benefits, of course, using CORBA, like being
>able to automatically locate a service, selective security access, etc.
>
>Dave
>
>
>>John
>>
>>
>>
>>_______________________________________________
>>Openmed-developer mailing list
>>[EMAIL PROTECTED]
>>http://lists.sourceforge.net/lists/listinfo/openmed-developer
>
>David W. Forslund [EMAIL PROTECTED]
>Computer and Computational Sciences http://www.acl.lanl.gov/~dwf
>Los Alamos National Laboratory Los Alamos, NM 87545
>505-665-1907 FAX: 505-665-4939
>
Join 18 million Eudora users by signing up for a free Eudora Web-Mail account at http://www.eudoramail.com
_______________________________________________
Openmed-developer mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/openmed-developer
Computer and Computational Sciences
Los Alamos National Laboratory
505-665-1907
