You assume the worst of me. It seems that looking at actual
implementations of both 13606 and V3 will provide excellent experience data
for both groups. I know V3 implementations, and did not know many 13606
implementations, altho I do know one system that has several
implementations. So Bert, I would like your help.
Ed Hammond
|---------+------------------------------------->
| | Bert Verhees |
| | <bert.verhees at rosa.nl> |
| | Sent by: |
| | openehr-technical-bounces@|
| | openehr.org |
| | |
| | |
| | 10/16/2006 05:19 PM |
| | Please respond to For |
| | openEHR technical |
| | discussions |
| | |
|---------+------------------------------------->
>--------------------------------------------------------------------------------------------------------------------------------------------------|
|
|
| To: For openEHR technical discussions <openehr-technical at
openehr.org> |
| cc:
|
| Subject: Re: Antw: Re: Antw: Re: AW: HL7 templates/archetypes
|
>--------------------------------------------------------------------------------------------------------------------------------------------------|
Williamtfgoossen at cs.com schreef:
Good Point Ed,
Until now the list of actual OpenEHR implementation I have actually
seen working is 0
William,
I told before on this list, English is not even my second language, but I
do what I can to be understandable.
-------
Now to your "Good Point, Ed" (the famous G.P.E.)
I don't think this is fair, I explain why I feel like that.
Release 1.0 is just released this year, I am working with a group of people
on a commercially implementation. At this moment (for business-reasons) I
cannot go into further details, but we will within a few months.
If you need an implementation, contact me, I will lead you further, I can
get you to an implementation within a few months..
If you only want information, please wait for more press releases.
The concepts are very new, the learning curve is steep, So don't expect a
lot of implementations in short time, but, in my experience with complex
software like this, there is a point of critical mass after things will
really go fast. Also, I know about other implementations in beta phase.
When people involved are willing to provide information on this, you will
learn about it from them.
----------------
I worked on HL7 implementations, I cannot say that it is a pleasure. I
still work on code to create the messages Nictiz defined. They are very
complicated, a lot of its complexity cannot be used at all because most of
the providing GP-applications do not support that kind of
data-constellations. So lots of definitions just exist in vain, as a kind
of hobby for the designers.
Also, I guess that the dutch GP-systems will need maybe three to five years
to get an error-free partially implementation. When you see how long they
needed to implement the much more simplier Edifact (medeur) message from
the Erasmus university. They started at that time with 1500 errors in each
Edifact message.
Stupid errors, like sexe: Islam, religion: O-neg Blood type: male. They had
lots of problems to distinguish the + and ' and other characters which have
special meanings in Edifact. It took more then three years to let them
prouce/implement edifact messages of an acceptable level of errors.
I am not a genius, but I wrote a Medeur testing tool, a medeur-readable
report generator and other medeur tooling in a few months, on my own, they
are still in use..
Now they have to build a system for HL7, which is much much much more
complicated as Edifact. (Medeur). I do not expect them to succeed to
implement the HL7 v3 message on an acceptable level error-free level before
2010. Say, it will be 9 years after Nictiz started defining the PRICA-DMIM
(primary care) (which also took 4 years and 50 million Euro, but being
fair, it was not only the DMIM, also the SOAP headers and some more things
had to be done. Maybe there were other DMIM's too they defined. I know one
more, BEMO (medications)).
So, At this moment there is no application in my knowledge which implements
a trustworthy HL7 v3 Prica message-generation or -interpretation.
On the other side I am very confident that before 2010 there will be also
several systems in the Netherlands running OpenEhr.
-------
Politicians and people with not much knowledge of what is really happening
keep on pointing to the Proof Of Concept which took place last summer, and
try to gain some good hope from that. Many people have a wrong idea of what
this PoF really was about (in GP-communications).
The PoF did not test any GP System. It was really only a
network-service-test, not a GP-information system-test.
I know of a company which name I will not mention, which received, really a
lot of money (was it 100.000 Euro) to receive a request for a HL7 message,
and in reply send a message, which was in every detail before prepared (a
simple Perl script of fifty lines could do the job). No life GP-information
system was involved in this test, it was only an network SOAP test.
OK, the SOAP headers are OK, now the GP systems.
These are things you don't read in the newspapers, you don't hear talk
about in management-meetings.
Thanks
Bert _______________________________________________
openEHR-technical mailing list
openEHR-technical at openehr.org
http://www.chime.ucl.ac.uk/mailman/listinfo/openehr-technical
_______________________________________________
openEHR-technical mailing list
openEHR-technical at openehr.org
http://www.chime.ucl.ac.uk/mailman/listinfo/openehr-technical
