Re: Does Codd's view of a relational database differ from that ofDate&Darwin?[M.Gittens]

Alfredo Novoa wrote:
> On Wed, 22 Jun 2005 19:12:20 GMT, Jan Hidders
> <jan.hidders_at_REMOVETHIS.pandora.be> wrote:
>
>

>>>>The words "procedural" and "imperative" are in this context largely 
>>>>synonymous.
>>>
>>> 
>>>They are largely confused, but not synonymous.
>>>
>>>http://foldoc.doc.ic.ac.uk/foldoc/foldoc.cgi?imperative+language
>>
>>The context in which the term was used was query languages, not update 
>>languages.

Alexander's words you reacted to where:

|
|>>Yet, I agree that FDM queries are still different from purely
|>>declarative queries.

> On the other hand a Daplex query reads the current state of variables
> and that is imperative programming.

No, it evaluates an expression for each element in a certain set of binding without the possibility of those expressions having side-effects that can influence each of those evaluations. That's called set comprehension and if you don't believe me then I'm sure a few experts in comp.lang.functional will be more than willing to explain to you why that is not imperative.

• Jan Hidders