Re: SUPPORT FOR DECLARATIVE TRANSITION CONSTRAINTS
From: paul c <toledobythesea_at_oohay.ac>
Date: Mon, 20 Sep 2010 21:42:15 GMT
Message-ID: <XyQlo.855$u9.454_at_edtnps82>
> This constraint is a firewall to protect supervisors from making
> stupid mistakes. The only way LBRDATE can be wrong is if the
> supervisor--not the employee--inserted an activity with the wrong
> date. Employees key information into barcode data collection
> terminals, and LBRDATE is set to the date the employee's shift started
> by the application that polls the terminals. Since the employee
> actually performed a labor activity on that date, it doesn't make
> sense for the supervisor to claim that it was performed on a different
> date. It does make sense for them to be able to fix their own screw-
> ups, however. If they insert a labor activity with the wrong date,
> they have to delete it and insert another one.
Date: Mon, 20 Sep 2010 21:42:15 GMT
Message-ID: <XyQlo.855$u9.454_at_edtnps82>
On 19/09/2010 6:14 PM, Brian wrote:
> On Sep 19, 6:25 pm, paul c<toledobythe..._at_oohay.ac> wrote:
>> On 18/09/2010 6:42 PM, Brian wrote: >> ... >> >>> Transition constraint #4: >>> LBRDATE can't be different. >> >>> Business rule implemented: >>> - The date that a labor activity was performed does not change. >>> ... >> >> What can this possibly mean? eg. different from what? change from what? >
> This constraint is a firewall to protect supervisors from making
> stupid mistakes. The only way LBRDATE can be wrong is if the
> supervisor--not the employee--inserted an activity with the wrong
> date. Employees key information into barcode data collection
> terminals, and LBRDATE is set to the date the employee's shift started
> by the application that polls the terminals. Since the employee
> actually performed a labor activity on that date, it doesn't make
> sense for the supervisor to claim that it was performed on a different
> date. It does make sense for them to be able to fix their own screw-
> ups, however. If they insert a labor activity with the wrong date,
> they have to delete it and insert another one.
That's what I would have hoped (delete and insert, aka negation and union), even from chronic under-simplifiers. But then I have to ask what would it mean if there were no such 'transition constraint'. Could the activity 'change' by any other means than delete and insert/negation and union? Received on Mon Sep 20 2010 - 23:42:15 CEST