On Wed, 2006-09-20 at 04:33 -0700, Larry wrote:
> We had to stop using SS# because of privacy concerns.  Can you remember
> a time when everyone used SS# for things like student numbers?  I can.
> Now, it's sacrilegios.

The banking and brokerage industries still use it.  The medical and
insurance industries (and countless others) always ask for it.  They may
not use them as primary keys officially but I'd be surprised if they
don't have a UNIQUE index on it.

> I wonder about the guy who thinks sequential keys are almost
> sacrilegious.  What does he do in situations where no natural key
> exists?... like a person object?

I don't know.  I'll limit myself to quoting a recent rant of his, in
response to someone who was asking for support of "integer indexes".

---
"I don't know what you mean here.  Possibly you are looking for 
auto-increment columns? ... A knowledgeable and disciplined 
database designer will never use them lightly, but in the rare case
where a surrogate key is genuinely unavoidable they'll be neat enough.
My objection to these is the past was that they discourage designers
from learning about the serious but non-obvious problems surrogate keys
hide.   I concede that battle is entirely lost, so what the hell?  If it
sells product..."

"Another problem with sequential keys in general--apart from the fact
that users invariably start to attribute unexpected meaning to them--is
the effect they have on the performance of highly concurrent systems
when used with B-trees."
---

These are not entirely unreasonable points.  In any case, I suggest that
if anyone wants more they can visit comp.databases.theory for regular
flame wars on the subject.

Joe


--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"pylons-discuss" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/pylons-discuss
-~----------~----~----~----~------~----~------~--~---

Reply via email to