Home -> Usenet -> comp.databases.theory -> Re: views of binary operations

Re: views of binary operations

From: paul c <toledobythesea_at_oohay.ac>
Date: Mon, 17 Jul 2006 14:36:09 GMT
Message-ID: <tfNug.200176$IK3.77245_at_pd7tw1no> 






Bob Badour wrote:

> Aloha Kakuikanu wrote:




>> Marshall wrote:


>>

>>> Aloha Kakuikanu wrote:

>>>

>>>> Marshall wrote:

>>>>

>>>>> Consider named views of binary operations on relations.

>>>>>

>>>>> Given a relational operator "op" and relation variables A and B,

>>>>> and a declaration of:

>>>>>

>>>>> r = A op B

>>>>>

>>>>> the language evaluates the expression "A op B" and assigns the

>>>>> result to r.

>>>>>

>>>>> However, if we declare this as a view, ...

>>>>

>>>> View is a named expression. r is a view.

>>>

>>> Is it necessarily named? Might we find use for an anonymous view?

>>

>> Well, the point is that the term "view" is redundant and SQL-ish.

>> Relations and relational expressions are the fundamental concepts, and

>> view is some bastardized idea.

> 
> I disagree. Views and snapshots are fundamental and are the source of 
> logical independence.





Not familiar with the term 'logical independence' as written - does it 
refer to logical-physical independence and/or lack of redundancy?



(I notice that Codd 1970 mentioned the word 'view' several times, I 
think once as a synonym for model and also in non-specific terms but not 
in language terms, say as a named relational expression or catalogue 
macro, ie. not in today's sense, if I've got it right.  As a 'macro', a 
view already defers some evaluation, if that's what Marshall S. is 
driving at.   Maybe that's not important, after all what we call 
relations today he called relationships then.)



p
Received on Mon Jul 17 2006 - 16:36:09 CEST












Original text of this message