Re: [Gnumed-devel] GnuMed Advocacy

[Karsten Hilbert]

>> At the same time i want to raise a cautionary voice on behalf
>> of the international GNUmed community. I would like to ask all
>> active developers to not make design and implementation
>> decisions based on Australian requirements and needs only.
>> This would greatly hamper GNUmed's applicability in other
>> countries which would be a great shame.

> No worries. We will not agree to anything that would impair gnumed's 
> international intentions.
That is good to know.

> We will not accept any funding that would 
> constrain us in that way either.
Maybe I can point out a difference here: It is IMHO in certain
case even acceptable to receive funding for doing particular things
now rather than later (such as implementing Australia specific
modules). I am fine with that. However, care should be taken
to not accept funding that is trying to restrict _what_ one
can do or _how_ one goes about it.

> Getting a *generic* industry independend drug database like the Australian 
> Medical Handbook would be invaluable even for other countries (after all, 
> Amoxycillin here in the same Amoxycillin as in Germany).
Absolutely.

> This is even more 
> valid for decision support integration.
Although in principle I totally agree these things may suffer
from a lot of "official" NIH-itis (Morbus Not-Invented-Here).
I do foresee at least major unease for German officials to say
"Ah, well, there's Australian backed guidelines in there, so
we are cool with that." However, this doesn't mean in any way
those guidelines shouldn't go in.

Maybe there could be two mechanisms: Type 1 guidelines being
tied into the software fairly deeply that encode common best
practice (Adrian Midgley's concept of "Penicillin vs.
Penicillamin" or warnings about pregnacny status during
prescription or "Mrs. Jones who is now here hasn't had her
3rd Hep B shot yet which was due last month !" - that
sort of thing). Type 2 guidelines being country or even point
of care specific things that plug into a generic guideline
engine. I know this is rather hard to do technically but it
would be a cleaner solution.

As you all know there's a few projects out there trying to
define meta standards on how to encode clinical guideline
knowledge. Also, Ian once suggested a particular way of going
about this with regard to drug side effects/interactions,
mainly.

> The only thing that probably will happen is that the Australia specific 
> gnumed extension modules will get a temporary priority in order to get at 
> least one working (showcase) version out soon.
This is alright and will likely increase general momentum
according to one of the findings of The Cathedral and the
Bazaar which says that most successful OS projects start from
a codebase that already does something (which we nearly have
already :-)

Karsten
-- 
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346