Re: circular relationships ok?

From: Alexandr Savinov <>
Date: Thu, 02 Mar 2006 13:33:41 +0100
Message-ID: <4406e626$>

Roy Hann schrieb:
> "Alexandr Savinov" <> wrote in message
> news:4406cd86$

>> Roy Hann schrieb:

>>> Rubbish.  I refuse to believe; I don't refuse to be persuaded.
>> Nobody is going to persuade you.

> Well evidently nobody is going to try anyway.
>> Only you can do it (if you want to).

> I am very good at persuading myself of things. Mostly they are wrong or
> insane. That's why I try to keep them to myself until I have formed an
> argument that I hope would persuade a skeptic.
>> I have not made any assertions

> I believe we may be encountering a language barrier here. Since (to my
> shame) I speak only English I guess we're not going to get any further with
> this because I can't express what I said any differently.
>> You are already nowhere because you lost the focus of this thread. Here
>> is again the problem (as far as I understand it): how to get rid of
>> cycles in modeling relationships. Do you have a solution? Do you have
>> questions concerning my solution?

> I guess I do have a question actually. Why would I ever want to get rid of
> a cycle? (What fundamental purpose is served by that? What fundamental
> damage is not done by that?)
>> Whether you are interested or not is your own problem and (again) I am
>> not going to teach you what direction in data modeling to choose. I can
>> only *inform* you *if* you have an interest.

> I think you are promoting an unnecessary idea (for fun or profit, I don't
> care). For various reasons I dislike seeing unsubstantiated assertions
> about mysteriously beneficial database design methodologies. On the other
> hand, I am keenly interested in good ideas that solve real problems
> properly. I am quite ready to read about them, right here, right now,
> because I could be wrong. Please, get me interested. One paragraph.

If you are really interested in new ideas (although I seriously doubt) then you might want to read a paper about a new programming paradigm, called concept-oriented programming (COP):


Shortly, a new programming construct is introduced, called concept. It is a combination of one object class and one reference class. It generalizes conventional classes as defined in OOP. It is not directly connected with this forum, however, it complements the concept-oriented data model (COP) and hence can be useful for understanding how data modeling and programming are connected. COP is formulated as a method of programming however it describes thing which are important for data modeling, for example, how data elements are represented and how they are accessed.

And please, do not ask me to force you to believe in this (or any other) theory. If you like it and want to know more then ask questions. If you do not like it and want to say why then post your opinion or ignore it.

Received on Thu Mar 02 2006 - 13:33:41 CET

Original text of this message