Re: Objects and Relations

On Feb 10, 12:25 am, "JOG" <j..._at_cs.nott.ac.uk> wrote:
> On Feb 9, 11:58 am, "David BL" <davi..._at_iinet.net.au> wrote:
>
>
>
>
>
> > On Feb 9, 12:09 pm, Gene Wirchenko <g..._at_ocis.net> wrote:
>
> > > "David BL" <davi..._at_iinet.net.au> wrote:
>
> > > [snip]
>
> > > >All the above comes down to definition. Your usage of "model" is
> > > >different to mine. I restrict it to the narrow sense of the
> > > >representation of knowledge (in the form of attributes and
> > > >relationships) about external entities. I define "simulation" to be
>
> > > By that, a clay model of an automobile would not be a model. I
> > > think you have an overly-restrictive definition.
>
> > The subject of this thread is a comparison of OO and RM. Does it
> > matter if it doesn't encompass clay models? How is it too restrictive
> > for database systems, RM or cdt?
>
> > > >about creating a working machine that happens to *mimic* reality with
> > > >no intention of storing knowledge about external entities. Are you
> > > >saying such a distinction doesn't exist? If so then we disagree. If
> > > >it is merely over what the words "model" and "simulation" mean there
> > > >is nothing interesting to discuss.
>
> > > A simulation is a model.
>
> > > A model is a restricted representation of something. The
> > > simplifications depend on what the model is to be used for. If a
> > > model for aerodynamic testing, the shape of a car would be modelled.
> > > The sound system probably would not be in the model (though the aerial
> > > might be).
>
> > Yes according to your definition a simulation is a model.
>
> Few things. First when you are trying to express something to people
> in a certain newsgroup it is always beneficial to use the terminology
> specific to that domain if one wants to be listened to. Out of his own
> goodwill, mAsterdam maintains an extremely good glossary for the
> newsgroup, and that really should be the first port of call.
>
> Second, it is worth noting that it has been written that Codd
> regretted naming his theories the relational "Model", very much
> because of this sort of confusion. RM is about recording and
> manipulating propositions, and not really about "modelling" at all.

I think "model" means different things to different people. Some people would say RM is about modelling. I find the question rather boring.

>
> Third, I am still not clear at all what the aim of this thread is. It
> all seems a bit like a lot of handwaving in search of an actual point.
> A point that I am yet to hear succinctly put.

I shall try to summarise my conjecture...

1. For the purposes of building a system using OO and/or RM it is useful and meaningful to distinguish entities that are inside versus outside the abstract machine.
2. OO is inherently about building the abstract machine. Object decomposition is about decomposing the machine into smaller objects that have identity, state and behaviour defined with respect to the abstract machine. The identity and behaviour must not be confused with an entity outside the abstract machine. At best an object can mimic an external entity in the manner of a simulation. It certainly should not be regarded as merely for the purpose of knowledge representation because that is in conflict with objects having there own independent identity, state and behaviour.
3. RM is inherently about recording knowledge about entities outside the abstract machine in the form of ground facts or propositions.

The conjecture is simple yet powerful and therefore should be easy to falsify. For example it suggests that RM is a poor choice for string processing. It suggests that OO shouldn't be concerned with storing knowledge about real employees.

These points were clearly stated in the original post. The claims are not entirely awe inspiring, revolutionary or heretical. I have no idea why this thread has gone on for so long. It all seems rather pathetic. It has left a bad taste in my mouth and my opinion of the newsgroup has taken a dive. I have little respect for the "cult" nature of the group - complete with a so called "gate keeper", the lack of intellectual honesty, the repeated insults, the arguments over definitions, repeated use of metaphorical argument, repeated use of meta-physical argument, general inconsistency, arrogance etc.

This has been done to varying degrees by different members of cdt. For your part Jim, you have some admirable qualities. I also found the discussions with Marshall to be worthwhile.

Anyway, I've come to the conclusion I can do better things with my time. Received on Sat Feb 10 2007 - 02:23:46 CET