Re: SUPPORT FOR DECLARATIVE TRANSITION CONSTRAINTS

On 20/09/2010 12:44 PM, Brian wrote:
> On Sep 20, 12:42 pm, paul c<toledobythe..._at_oohay.ac> wrote:
>> On 20/09/2010 1:40 AM, Brian wrote:
>> ...
>>
>>> Please go back to school and take an introductory course in logic.
>>
>>> Tuples are supposed to represent propositions. Propositions refer to
>>> objects in the universe of discourse. In the context of database
>>> updates, it is a gross oversimplification to assume that the universe
>>> is nothing but a collection of arbitrary objects that are independent
>>> of time....
>>
>> Actually, much, if not all, of db and computing in general depends on
>> gross oversimplification, no matter that some theorists might prefer the
>> more expensive word, abstraction. But there's nothing gross about it if
>> the resulting application is deemed useful.
>>
>> In the context of this particular application it's an out-of-context
>> non-sequitur to mix up 'the universe' with 'universe of discourse', what
>> I would call an act of high mysticism.
>
> Excuse me, but are you trying to say that propositions don't refer to
> objects in the universe of discourse?
> ...

Actually, I was trying to say that gross undersimplification is irrelevant and wasteful and mysticism to boot. It's not necessary and is unhelpful to encumber a practical example of a real or supposed problem with loaded terms like 'object'. The propositions that are (partially) represented by tuples are simply the statements people wish to record or assert in an abstract but formal way. Doctrines or taxonomies of 'objects in the universe of discourse' aren't needed, no matter how important they might seem to sound. Regardless of the merits of your suggestion, it's enough for the example problem to talk only of 'work orders', 'labour activities' and such. Received on Mon Sep 20 2010 - 23:31:25 CEST