Re: Tools for automated normalization ?

From: mAsterdam <mAsterdam_at_vrijdag.org>
Date: Tue, 03 May 2005 00:17:39 +0200
Message-ID: <4276a6fd$0$150$e4fe514c@news.xs4all.nl>

tomjbr.10216233_at_bloglines.com wrote:

>>I think this is because nobody actually normalizes the way the
>>theory is explained, that is go from relations in 1NF, check them
>>for anomalies and subsequently normalize this set into 2NF etcetera.
>>The only way I've ever seen the normal forms used is to
>>validate/critisize a design proposal.

>
> I think it is very suprising that noone yet have implemented it,
> considering that for example sourceforge.net currently have about
> 100.000 registered open source projects.

In the early 1990's several case tools had 'normalization support'. The feature was'nt worth the effort to the users. No demand for it.

> I mean, if programmers can figure out that many kind of applications or
> components to create, then at least I am surprised that noone yet have
> come up with the idea to implement the normalization algorithm,
> considering how very many people are interested in using normalized
> relational databases.
>
> However, even though I can not understand it, it seems to be the
> reality and I guess you are right, since nobody seems to be aware of
> any available implementations of such an algorithm.

I don't remember if it was IEF (Inormation Engineering Facility) or IEW (Information Engineering Workbench) which had quite exentensive support, but I do remember this support just did not help the design process at all. Received on Mon May 02 2005 - 17:17:39 CDT