Re: [Gnumed-devel] Drug Progress

[Karsten Hilbert]

Horst,

> The MIMS data in it's original form does not lend itself to decision 
> support as we envision it.
This may very well be true but I'd rather offer a non-perfect
existing useful solution than a perfect non-existing solution.
Oh, and I am fine with crapping code later on when the better
things become available. Even my own (code, that is). But
until that time I *also* support that we'd better use what's
already there. And /already there/ means Hilmar+me+AMIS and/or
Richard+MIMS or even Bob Shepherd+Alex Caldwell+Multum or
Elizabeth Barham(?)+US-NDA.

> I do think that MIMS will play a role for us in the foreseeable future, 
OK, I almost got the impression you didn't, anymore.

> but the role for it as I see it is a complementary one, as an optional 
> source for our generic drug database for those drugs where we don't have 
> (enough) information from other sources.
Agreed.

> But we must not limit ourselves to their data model!!!
Certainly not.

> And we must not commit the mistake of having more than one
> API for pharmaceutical reference.
Certainly not. But this does not preclude several _backends_
for disparate data sources :-)

> Hence, MIMS dta can only be used by massaging it into our own 
> drugref based database.
Give me a document on how to setup the drugref based database
as a read-only drug info reference service and code to use it
as such (not just a proof-of-concept, I know the concept
works, it needn't be proven to me) and I will start using that
any day. Or, provide a remotely accessible database on
drugref.org (==gnumed.net) and some code to access it as a
service. I shall immediately refrain from thinking about
a MIMS specific backend to the general drug info reference
API.

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