Re: Guidelines to a decent support of surrogate key implementation

On May 24, 2:33 pm, "David Cressey" <cresse..._at_verizon.net> wrote:
> "Cimode" <cim..._at_hotmail.com> wrote in message
[Snipped]
> I'm relieved.
>
> > I used physical mechanism for consensual communication
> > purposes but I thought internal row adressing when getting physical.
> > I observed that opinions diverge on that perspective.
>
> I think that an internal row address identifies and locates a particular row
> (the container). I think that a surrogate key identifies (by content) the
> data in a particular row. I don't think the two ideas are substitutes for
> each other, in a context where these two are visible.
>
> In a context where the two are transparent, I don't see why the phrase
> "surrogate keys" needs to be used at all.
Agreed. But that's the conclusion not the thinking process (I hope you understand that the form of the subject question is intentional). I thought about using a term people can relate to to trigger thought for *not necessarily* educated audiences.

> > Hope this clarified. I am trying to expose several point of views
> > here to allow people to expose their perspective on that matter.
>
> I also want to see several points of view...
Then I suggest you follow the frame proposed (PRO/CONS). NG exchanges are difficult (Hoping I did get my point across). I believe it is important to communicate in a concise manner with a specific frame. But that's my opinion. Received on Thu May 24 2007 - 14:46:08 CEST