Re: [Gnumed-devel] next GNUmed release

[Karsten Hilbert]

On Sun, Sep 18, 2011 at 01:30:04PM -0700, Jim Busser wrote:

> > I would prefer us to resolve the clin.vaccine schema question in which I 
> > proposed that
> > 
> >     id_route
> > 
> > be removed from the table.

On careful consideration I think you have presented
sufficiently (to me) compelling arguments that - in its
present form - both .id_route and .is_live at least *can*
present clinical misinformation unavoidable by design such as:

        - is_live can be different among ingredients of a combi vaccine
        - is_live can be unknown
        - id_route can be multi-valued within a vaccine
        - id_route can be unknown
        - id_route can be different from actual use case by case

none of which is currently properly represented.

That is to be avoided.

I have entirely disabled support for both in the GUI. The
columns still exist in the backend awaiting better
resolution later on. Both are NULLable and DEFAULTing to
NULL, too, now.

Karsten
-- 
GPG key ID E4071346 @ gpg-keyserver.de
E167 67FD A291 2BEA 73BD  4537 78B9 A9F9 E407 1346