Re: relative complement?

On 04/04/2011 5:37 AM, Erwin wrote:
> On 4 apr, 10:54, David BL<davi..._at_iinet.net.au> wrote:
>> On Mar 19, 7:47 am, paul c<anonym..._at_not-for-mail.invalid> wrote:
...
>> I accept the idea that update operators on views can be defined to be
>> unambiguous, dispelling a common argument against view updates such as
>> join deletes and union inserts.
>>
...
>
> (1) If views are used as a means for achieving logical data
> independence, then this is a non-argument, because users will then not
> be aware of any base relvars, by definition of what logical data
> independence means, and it then follows necessarily that any
> consideration as to whether or not they "should have update authority
> on the base relvars", is itself merely a function on whether they do
> have update authority on the virtual relvars referring to such base
> relvars !
> ...

I have the feeling that the term 'logical data independence' is just an idiosyncratic way of re-stating logical consistency and that it cropped up as a result of people fastening on views as some kind of logical outgrowth of relational theory rather than a mere efficiency of recording. With the latter viewpoint a more important question might be whether a seemingly less efficient recording leads to a clearer understanding of consistency/lack of contradictions, eg., instead of recording a join, record (somehow) all the individual propositions implied by the join including the constraints the join implies (eg., so-called inclusion dependencies) then imagine what must result in order to retract one of them.

I don't have the skills to express this very formally but my casual attempts lead me to think that most of the examples people bring up (including me) are incomplete or simplistic because they neglect to include all the pertinent propositions, aka, loosely, all the pertinent relations. Eg., if there is a table of suppliers, some of which may also be customers, I'd say recording all the propositions implied would require at least four tables, eg., suppliers, customers, constrained suppliers, constrained customers, maybe more for all I know. Whatever steps a view mechanism takes (or doesn't take) to replace a relation value shouldn't contradict the less efficient recording and I don't think one can say a resulting 'view' value is ambiguous unless it can be shown that some less efficient recording it can logically be paired with is also ambiguous.

'Security' views seem to me to be a red-herring. The 'secured database user' is akin to the 'designated idiot' recurring fixture in other human schemes who has his own separate logic, ie., from others' viewpoints, his own distinct unrelated propositions. Received on Mon Apr 04 2011 - 17:14:17 CEST