[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Gnumed-devel] (no subject)
From: |
syan tan |
Subject: |
[Gnumed-devel] (no subject) |
Date: |
Fri, 17 Oct 2003 14:42:46 +1000 |
User-agent: |
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.3) Gecko/20030313 |
What the perssimists are saying:
from The_Myth_of_self-describing.pdf
The GP system might store demographics as:-
firstname
lastname
middle initial
street number
suburb/town
postcode
employment status
No amount of XML formatting is going to make these systems
interoperable. The business
logic at each end needs to change. They both need to conform to a common
conceptual
model. Even though the GP might be able to, the GP system cannot simply
"understand"
XML messages of the form :-
<name>"Rev. Eric Browne"</name> <address_line_1>"c/o
Bournemouth Caravan Park"</address_line_1> <address_line_2>
"Camden South"</address_line_2>
<employment_category>"baker"</employment_category>
Sounds like a load of monopolistic, reactionary hogwash. If you can
create a specific parser that can parse the fields into the other
fields, do any searches necessary to fill in implied information (
Camden South + implied country -> postcode , baker -> employment
status = employed)
and also if the target system's record might have a extra info field for
raw data (e.g. source_raw_data) or even add another separate table (
e.g. translations tablename, pk_value, source, raw_text ) , than
such a parse should be able to exist in xml - e.g. one can write a
Turing machine in xsl ( see "Stupid things One can do with XML") , and
a Turing machine
is suppose to model any general purpose computer, then a parser should
be possible.
The argument should be - if everyone would give up the Tower of Babel
and just use OUR
reference model (openEHR), then heaven would be on earth, or something
like that.
(As a logical step, why don't we all start using Esperanto).