On Tue, Apr 26, 2005 at 06:36:31AM +0800, Syan Tan wrote: > you could stick boolean is_action, ref_action on clin_root_item > action ,action_type, action_intent tables , and voila , metamodel of actions > in gnumed , no? Well, hm, clin_root_item is intended to root a (shallow) tree of passive data items. It wasn't designed to represent actions and I'm fairly sure there would be shortcomings because of that.
I am also not so sure what recording actions this way buys us ? Once one goes into modelling actions/observations etc in all their flavours - which is at least on magnitude of quality beyond current GNUmed - one should go all the way and build upon an OpenEHR kernel. Karsten -- GPG key ID E4071346 @ wwwkeys.pgp.net E167 67FD A291 2BEA 73BD 4537 78B9 A9F9 E407 1346 _______________________________________________ Gnumed-devel mailing list [email protected] http://lists.gnu.org/mailman/listinfo/gnumed-devel
