Re: c.d.theory glossary -- definition of "class"

From: mAsterdam <mAsterdam_at_vrijdag.org>
Date: Thu, 24 Jun 2004 01:45:49 +0200
Message-ID: <40da162e$0$34762$e4fe514c@news.xs4all.nl>

Alfredo Novoa wrote:

>>>Most times it is a type.

> mAsterdam wrote

>>This is the us vs them explanation.
>>'us' (relational) database redefining the world,

>
> No, this is 'us' the "serious" people trying to translate the
> charlatans' pap into something whith a minimum of precision and
> rigour.

Ah :-) count me in.

>>them OO developers who don't even get their central
>>concepts clear.

>
> I am an OO developer, and I want to have my central concepts clear.
> I had big problems in the past when I confused everything following
> the charlatans.

Well, I am not an OO developer (anymore), but I have to deal with OO developers and DBA's,
and I very much feel the same way.

>>This does not help understanding across
>>paradigmatic boundaries at all.
>>On the contrary, it demarcates.

>
> It demarcates the serious guys and the charlatans.

Here we differ.
I think you are overreacting. The use of the word 'class' does not make you a charlatan.
I'ld almost say the putting it on a blindspot would - but I won't. I don't think you are a charlatan :-)

>>Behavior is a central aspect to OO. Any respectful
>>description of the term class should include behavior.

>
> Behavior is a very informal and fuzzy term. The behavior is determined
> by the semantics of the operators. This is a term we should drop in
> formal contexts.

Complex, yes. Dynamic, yes. Fuzzy, no.
Fuzzy here is the blurring of semantics and pragmatics. The behavior is determined by the *use* (as opposed to *meaning*) of the operators. It is the pragmatics, not the semantics (meaning). In formal contexts dealing with data at least this distinction should be clear.

>>So let's abolish all terms that have more meanings?

>
> Yes if we have alternatives with only one meaning.

Not if these disregard valuable notions.

>>There won't be many words left.

>
> A good thing :)
>

>>The term is used. It is used by a lot of people who use databases.

>
> We should help to reduce the number :)

Of people using databases? ;-)

>>It is used in relation to the use of databases.
>>It is important to have an understanding of what is meant
>>when 'class' is used in the context of databases.

>
> Many things. That's the problem.

Well, yes. But IMHO there is an essential, (dynamic) notion getting lost if we use 'type' as a would-be synonym.

>>Somebody who strikes it from the vocabulary simply won't
>>hear what is said when the term is used.

>
> I want to know what is intended to be said, that's why I ask for more
> precise terms.

So am I. Extending 'type' just blurs. It does not help. Received on Wed Jun 23 2004 - 18:45:49 CDT