From: | "Jeff Eckermann" <jeff_eckermann(at)yahoo(dot)com> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: artificial keys or not? |
Date: | 2005-04-25 15:34:54 |
Message-ID: | d4j2j5$i28$1@news.hub.org |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
"Scott Ribe" <scott_ribe(at)killerbytes(dot)com> wrote in message
news:BE8ED6E7(dot)21FC1%scott_ribe(at)killerbytes(dot)com(dot)(dot)(dot)
>> ..so the only question is:
>>
>> drop the the serials as PKs and just use short text handles? our
>> database size is arround 290 mb. there shouldn't be speed issues as long
>> as proper indexes exit, right? some of the conditions are dynamic
>> though, so there can't be an proper index in any case. any other
>> concerns?
>
> Someone (a manager) somewhere (probably marketing) sometime (when you
> least
> expect it) will demand that those short text names be changed, possibly
> for
> a completely irrational reason. I know there are still text books that
> claim
> that artificial keys are evil for some inscrutable reason, but hard
> experience has taught me to never, ever, under any circumstance, make a
> primary key out of data that comes from humans, and to be very suspicious
> of
> using data that will be visible to humans.
This is my experience also. But opinions vary, and I wouldn't be surprised
to see a further post from someone which argues the exact opposite. :-)
>
>
> --
> Scott Ribe
> scott_ribe(at)killerbytes(dot)com
> http://www.killerbytes.com/
> (303) 665-7007 voice
>
>
>
> ---------------------------(end of broadcast)---------------------------
> TIP 6: Have you searched our list archives?
>
> http://archives.postgresql.org
>
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2005-04-25 15:36:23 | Re: timezone() with timeofday() converts the wrong direction? |
Previous Message | Akbar | 2005-04-25 15:29:55 | installation as service failed on windows xp |