Re: [Gnumed-devel] Time for a major re-think in 2005 - opinions please.

[Karsten Hilbert]

> I feel that interfacing to a large number of proprietary APIs
> might be more difficult than implementing a direct access to
> some basic drug database. Drugref or some rudimentary form of
> curated substance list might be sufficient for the beginning.
A simple substance list would only be useful for selecting
for prescribing. It does not help any as a source of
information on drugs. Usefully supporting prescribing,
however, takes a lot more data that a simple drug list. So
does drug data browsing. But the latter is less involved so
easier to attain and a reasonable optional and actually useful
goal for 0.1.

> In the end, access to any type of drug information source
> should be transparent, i.e. there should be some sort of
> drivers for remote/local databases and commercial apps and a
> common middleware/GUI that adapts to the capabilities of the
> backend.
It sounded like Horst already has this.

> That's the basic idea of the drug browser that is
> already existing in the Gnumed and has basic functionality with
> the german AMIS database. All it needs to connect other
> backends is to write drivers for them (and overhaul the design
> a bit).
Does it offer the option to "drive" other DB frontends ? Eg.
jump the user to that frontend, perhaps calling up a given
drug, waiting for the user to finish whatever business she had
within that foreign frontend and continue working with the
results passed back by that frontend ?

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