Re: On view updating

From: mAsterdam <mAsterdam_at_vrijdag.org>
Date: Sat, 25 Sep 2004 14:12:17 +0200
Message-ID: <415560a1$0$43451$e4fe514c@news.xs4all.nl>

Costin Cozianu wrote:

>>> Since relvars are "variables" and the rest are computed values,
>>
>> Views are variables too. Virtual variables but variables. Although a
>> "closed view" has nothing of virtual to the user.

>
> There's no such thing as "virtual variables" in PL theory.So to avoid
> confusion by "my imaginary model" vs. "your imaginary model" kind of
> debate let's enumerate some invariants across all reasonable models
>
> - Views have a name
> - A view denotes an expression that can be evaluated or composed with
> other expressions.
> - views do not denote a "store".
>
> as opposed to views, tables, which also have a name:
>
> - denote a "store". A place in the global state holding a value that can
> be changed.

A store, which is completely hidden from the users of the table (encapsulated by the RDBMS). To the user, the table poses as store.

> Therefore, here you have your trivial distinction in all imperative
> languages between a "store" (variable, possibly aliased or non-aliased)
> which is your lvalue, and an expression.

Why could a view *not* pose as store?
Hey! You just invented the 'virtual variable'!

:-) Received on Sat Sep 25 2004 - 07:12:17 CDT