Re: SUPPORT FOR DECLARATIVE TRANSITION CONSTRAINTS

From: Brian <brian_at_selzer-software.com>
Date: Tue, 28 Sep 2010 12:42:14 -0700 (PDT)
Message-ID: <c57aeed5-c068-4221-9a92-c521fc282902_at_x42g2000yqx.googlegroups.com>


On Sep 28, 11:20 am, Erwin <e.sm..._at_myonline.be> wrote:
> On 28 sep, 16:05, Brian <br..._at_selzer-software.com> wrote:
>
>
>
> > No.  It isn't.  Only transition constraints can prevent a user from
> > rewriting history.  All a user needs to do to subvert any state
> > constraint that involves temporal attributes is to replace the history
> > of the object under consideration.
> > - Tekst uit oorspronkelijk bericht weergeven -
>
> All a user needs to do to subvert your stupid UPDATE transition
> constraints is not issue an UPDATE, but a DELETE+INSERT multiple
> assignment instead.

It's not the same thing. The DELETE+INSERT multiple assignment would have to meet all of the conditions for DELETE as well as all of the conditions for INSERT. Moreover, the user issuing the subversive DELETE +INSERT multiple assignment is telling a lie, whereas the user subverting a state constraint that involves temporal attributes is rewriting history. The system can't determine correctness, so it can't detect lies, but I think it should be able to detect a rewrite of history. Received on Tue Sep 28 2010 - 21:42:14 CEST

Original text of this message