On 08/11/2010 13:45, William E Hammond wrote:
> I appreciate all of the remarks that have been make thus far. I am
> responding because I think we might have some shot at being better. I
> think many of you tak pot-shots at HL7, and that's OK.
I want to clarify one thing: HL7v2 is an excellent standard in the
overall judgement of perfection versus pragmatism. It was developed with
large amounts of empirical evidence, and has slowly grown over the
years. I heard talk of 2.9 recently. It is relatively compact, solves
problems it addresses with a reasonable hit rate, and has a pretty good
cost/benefit ratio. Yes, it drives developers up the wall, and has all
kinds of warts. But the penetration and utility shows the real value. It
is no accident that the US and Australia and others make such heavy use
of it.
The problems we have today with HL7 I think are to do with having
entered into massive complexity while losing touch with evidence.
> Governmnets make decision. Governments fund efforts. To ignore
> governments would be foolish. Every country has an official government
> body that is responsible for standards - ANSI, BSI, DEN, AFNOR, others.
> Complexity causes collapse.
One of the main points I made on my blog about this was that in every
other industry I know of, the standards are created from fully
engineered products, usually created by companies, the military or
academia. In the case of toughened glass or mobile phone signalling, the
standards orgs are doing the right (more or less) job. In health
informatics, other than IHTSDO, they are on some other planet.
> You claim that ISO is flawed. Ballot is by standard, a only a few
> countries dominate. That obviously is not restricted to standards. Again,
> that's life.But what is a better solution? Shall we live with a decision
> making prosess in which a relative few people decide what is correct?
I think that today's world has shown us better solutions. Here are 2:
* IETF - largely built by dedicated academic, military and industry
people, produced an engineering framework on which most of our
modern communications work. The design work did not occur in
committees.
* the large open source projects, e.g. Linux and Apache to mention a
couple, and let's add Python, Plone etc, as Tim would no doubt do.
In both examples, a relatively small number of people do decide (in a
technical development environment) what is a correct solution to the
problem at hand. In the case of Linux, Linus Torvalds is famous for
being autocratic - but it works. This is life, not everyone is an
architect. The number of designers at BMW is but a tiny fraction of the
overall payroll. If it were any other way, we would have chaos. These
efforts then offer their output to the world at large, and the world at
large decides. Both IETF and the LAMP platform are massive successes.
That is because they did not decide on what was /correct for us/ - we
did that - we decided what /worked for us/.
The success and quality of the above efforts shows us just how flawed
building technical artefacts by the paper-based committee approach is.
We really need to have total reform, and as soon as possible, because
the wastage of having the best and brightest of the medical and IT
fields working in such a hopeless structure surely cannot be borne for
much longer.
> Can we create an open society that leaves some of the history and biases
> behind and find the best possible solution? Can we bring together the SDO
> organizations. I also have a problem that openEHR refuses to be an SDO.
I didn't know that openEHR had refused... I didn't even know that it had
been asked.
- thomas
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.openehr.org/mailman/private/openehr-technical_lists.openehr.org/attachments/20101108/633c93cd/attachment.html>
