"John S. Gage" wrote:
> >and so on. There are two desires in Europe: that it be HL7 v3 capable, and
> >that
> >it be (something close to) CEN compliant. In the US, people will want HL7
> >compliance.
>
> Yes, HL7 is the piece of the puzzle that corresponds to Linux in the
> operating system world. It's the arcana that open source must embrace to
> have an impact. Do you agree Thomas?
>
> John
I have certainly argued (and will continue to) that interoperability standards
and their implementations must be open source, to prevent closed source vendors
bulding solutions based on the standards, while a) making their own (unique)
implementation decisions, and b) silently changing/adding to the standard,
making their product not really standard. As far as I can see, the result of
this will always be fragmentation of the market, along vendor customer-base
lines, killing any hope of the interoperability that the standard was intended
for. Experience with HL7 v2, HTML, and all non-SGML word processors (i.e.
almost all WPs) bears this out (in the WP case, imagine that 20 years ago, some
OS collective started churning out something like an SGML kernel component,
that all WP vendors built their applications on top of. Voila: diverse GUIs,
features, with common file format. But it never happened...)
I think the place for closed source is for specialised solutions which can
compete - applications, special database bindings, etc; for interoperability,
which is by definition underpinned by cooperation, open solutions are the
desirable way to go. The other way to read it is that all software should be
open and probably free, since the support service is what buyers will pay for.
As far as HL7v3 goes, I don't believe that (m)any of the current vendors are
talking about open source. The problem for them is that they all have
businesses based on making stovepipes only, when they need to diversify and get
into applications as well, and make their HL7 implemenations free (or better -
just throw some money at a good OS project to build a rolls royce HL7v3
messaging system). I think if this does not happen, HL7v3 will repeat the
"implementation manual" nightmare of v2. This is not a criticism of HL7 per se,
just a comment on the potential mechanisms for development.
- thomas beale
--
..............................................................
Deep Thought Informatics Pty Ltd
Information and Knowledge Systems Engineering
phone: +61 7 5439 9405
mailto:[EMAIL PROTECTED]
Health Informatics - http://www.gehr.org/
Community Informatics -
http://www.deepthought.com.au/ci/rii/Output/mainTOC.html
..............................................................
