Re: pervasiveness of surrogate (also called synthetic) keys - Mailing list pgsql-general

From Craig Ringer
Subject Re: pervasiveness of surrogate (also called synthetic) keys
Date
Msg-id 4DBF8A33.4080003@postnewspapers.com.au
Whole thread Raw
In response to Re: pervasiveness of surrogate (also called synthetic) keys  (Greg Smith <greg@2ndQuadrant.com>)
Responses Re: pervasiveness of surrogate (also called synthetic) keys  (Rob Sargent <robjsargent@gmail.com>)
Re: pervasiveness of surrogate (also called synthetic) keys  (Merlin Moncure <mmoncure@gmail.com>)
List pgsql-general
On 03/05/11 11:07, Greg Smith wrote:

> That doesn't mean you can't use
> them as a sort of foreign key indexing the data; it just means you can't
> make them the sole unique identifier for a particular entity, where that
> entity is a person, company, or part.

Classic case: a database here has several tables indexed by MAC address.
It's used for asset reporting and software inventory.

Problem: VMs generate random MAC addresses by default. They're not
guaranteed to be globally unique. Collisions have happened and will
probably happen again. In this case, it wasn't a big deal, but it just
goes to show that even the "obviously" globally unique isn't necessarily so.

--
Craig Ringer

pgsql-general by date:

Previous
From: John R Pierce
Date:
Subject: Re: pervasiveness of surrogate (also called synthetic) keys
Next
From: Rob Sargent
Date:
Subject: Re: pervasiveness of surrogate (also called synthetic) keys