Re: Extending my question. Was: The relational model and relational algebra - why did SQL become the industry standard?

Lauri Pietarinen wrote:
>jan.hidders_at_REMOVE.THIS.ua.ac.be (Jan Hidders) wrote:
>>
>> Well, your specific example involves recognizing that duplicate
>> elimination is not necessary in the view. That was something that was
>> researched as (relatively) recently as 1994:
>>
>> http://www.informatik.uni-trier.de/~ley/db/conf/icde/PaulleyL94.html
>
>The subject has been covered even earlier in a paper by Hugh Darwen
>published in Relational Database Writings 1989-1991.
>The name of the paper is "The Role of Functional Dependence
>in Query Decomposition". It provides an algorithm for
>deriving all candidate keys from a relational expression
>(or view).

That's a different subject because he derives the keys *after* the duplicates have already been eliminated. But what you want to know is if there is a candidate key *before* you eliminate the duplicates.

• Jan Hidders