Re: Constraints and Functional Dependencies

Walt wrote:

> "Marshall" <marshall.spight_at_gmail.com> wrote in message
> news:1172382132.346588.167980_at_s48g2000cws.googlegroups.com...
>

>>On Feb 24, 2:08 pm, "Walt" <wami..._at_verizon.net> wrote:
>>
>>>"Marshall" <marshall.spi..._at_gmail.com> wrote in message
>>>
>>>news:1172333601.148573.19370_at_v33g2000cwv.googlegroups.com...
>>>
>>>>On Feb 24, 6:41 am, paul c <toledobythe..._at_oohay.ac> wrote:
>>>>
>>>>>Marshall wrote:
>>>>>

> Natural joins are easier to express, and easier to resolve. It depends on
> what metadata is used to set up the natural joins, I guess. But if you
> always use the primary key for reference, it strikes me that the resolution
> of natural joins has a smaller search space.

The algebra defines a natural join on the names of the attributes and not on the declared keys. Thus, declaring a primary key has no effect on the search space. Received on Tue Mar 06 2007 - 16:42:58 CET