we are developing GNUmed, a medical practice management
application running on PostgreSQL (you want your medical
data to be hosted by something reliable, don't you ;-) We
are putting out our first release sometime in the next two
The idea is to name the production database "gnumed0.1" for
version 0.1 (gnumed0.2 etc for upcoming releases). I do
realize the "." may force me to quote the database name in,
say, a CREATE DATABASE call.
The rationale is that when going from 0.1 to 0.2 I can
simply create the database "gnumed0.2" and populate it with
tables all the while gnumed0.1 is still up and running
serving users. Downtime would only be needed to actually
migrate the data (which, of course, is the bigger time sink
- unless we use Slony which may or may not possible due to
schema changes). The other advantage is that I can instruct
my doctors to "startup the previous client version if
something breaks at 3:30am and call tech support in the
morning". I *am* a doctor. I have personally been in that
very situation with commercial electronic medical record
applications. Which is part of why I am participating in
GNUmed in the first place.
Now, my question is whether I am failing to see the
*dis*advantages that may be looming from such an approach.
Anyone has any helpful thoughts on this ?
Karsten Hilbert, MD
GNUmed i18n coordinator
GPG key ID E4071346 @ wwwkeys.pgp.net
E167 67FD A291 2BEA 73BD 4537 78B9 A9F9 E407 1346
pgsql-general by date
|Next:||From: Adam O'Toole||Date: 2005-07-13 16:05:17|
|Subject: Re: INSERT INTO from a SELECT query|
|Previous:||From: Greg Patnude||Date: 2005-07-13 15:48:18|
|Subject: Re: Temp tables...|