Re: pervasiveness of surrogate (also called synthetic) keys

On 04/28/2011 11:44 AM, Andy Colson wrote:
> On 4/28/2011 12:29 PM, Jim Irrer wrote:
>> A colleague of mine insists that using surrogate keys is the
>> common practice by an overwhelming margin in relational databases and
>> that they are used in 99 percent of large installations. I agree that
>> many
>> situations benefit from them, but are they really as pervasive
>> as he claims?
>>
>> Thanks,
>>
>> - Jim
>
> I dont see how you could know unless you went to all the "large
> installations" and asked. But since its a good idea, and you "should" do
> it that way, and because I'm pessimistic, I'd say only 5% of RDB users
> do it that way.
>
> Oh! Joke: Why do DB Admins make better lovers? They use surrogates!
>
> Anyway, I'm not a large install, but I use em. That's gotta count for
> something.
>
> Really, how could you count? Was there a poll someplace? Ask for some
> data. Otherwise seems like BS to me.
>
> -Andy
>

Hm, I get the feeling that only the good folks at Hibernate seem to
think using a "natural key" is the _only_ way to go.