Re: Microsoft and the two great blunders

From: Costin Cozianu <c_cozianu_at_hotmail.com>
Date: Tue, 23 Dec 2003 19:03:56 -0800
Message-ID: <bsavog$bg13t$1_at_ID-152540.news.uni-berlin.de>


Joe "Nuke Me Xemu" Foster wrote:
> "Costin Cozianu" <c_cozianu_at_hotmail.com> wrote in message <news:bs9ob2$a1qdq$1@ID-152540.news.uni-berlin.de>...
>
>

>>Alfredo Novoa wrote:
>>
>>>Nonsenses
>>
>>Bullshit

>
>
> Unless the relation holds all possible values of a type,
> the relation cannot be considered equivalent to the type
> itself.

Well, if you want to hold to the relational orthodoxy, you ought to call those relvar. But to make it clear we'll call them table, shall we ?

Now who said that relation should be considered to be equivalent to the type itself ? Whatever that type might be.

Now let's further consider that the table EMPLOYEE stores all the tuples that have the type EMPLOYEE%ROWTYPE (using SQL terms, i.e. the type of the tuples).

I'm waiting for someone to make the case why would that be such a big deal ?

> Relations are sets and thus values, not types.

Quite nice. But according to D&D types are sets of values. Well, then a relation is a set of value.

According to proper type theory, a type is much more, of course.

According to E/R theory, that seems to be the source of inspiration for the powerpoint incriminated here for no reason, we talk about an Entity Type, which is identified by a name and a set of attributes (whereas an attribute is identified by its name, and has a type).

> The difference is pedantic only until ignoring it trips
> somebody up, and this has happened.
>

How it happened, where it happened, how often it happened ?

For lack of better things to do, people think that if they throw Microsoft and a couple of relational orthodoxy together, they've got a valid point.

Oh, well. Cheers,
Costin Received on Wed Dec 24 2003 - 04:03:56 CET

Original text of this message