Re: NIAM (was: Just one more anecdote)

On Fri, 12 Aug 2005 02:33:14 +0200, mAsterdam wrote:

(snip)
>I was so lucky to have been attending a session where
>professor Nijssen convincingly argued that the result
>would necesarily be a set of relations in 5NF without
>ever mentioning normalization in the process which
>lead to the set.

Hi mAmsterdam,

Yup. That's one of the advantages of using NIAM (or ORM, for that matter). Of course, it has it's disadvantage as well - I can churn out fully 5NF designs without much trouble, but please don't ask me to explain the difference between 2NF and 3NF! <g>

>I wonder: if we would introduce intervals as first class
>scalars would we wind up with Date, Darwen & Lorentzos'
>6NF without the hassle? I gave it some thought.
>I think so, but I can't prove it.

Do you have some links to relevant information about 6NF? I googled a bit, but what I found was more confusing than enlightening.

Maybe I should keep shut until I know more, but it strikes me as strange to define a specific interval scalar type for temporal data, but not do the same for non-temporal data. The world is full of rules such as "buy 20-50 pieces and get 5% off; buy 51-100 and get 7% off; but 101+ and get 10% off". Or "tax rate is 35% for the first € 35.000, 50% for € 35.001 up to € 80.000 and 60% for € 80.001 and up". Why should temporal intervals be treated different from other intervals?

One thing I saw when googling for 6NF gave me the impression that Date, Darwen & Lorentzos propose to define 6NF such that you can't take any column from a relvar without leaving the remaning data meaningless. If that is indeed what they proposed, then they have only coined a new term for what Halpin and Nijssen call "elementary fact types" - and a 6NF database design would be almost equal to a NIAM or ORM data model.

>
>> It's too bad that Universele Informatiekunde (Nijssen, 1993; published
>> by PNA Publishing BV) was never translated. Despite the many errors it
>> admittedly has, it's still a very good book.
>
>Is it still available in dutch?

According to the publishers' web site
(http://www.pna-publishing.nl/pagina.asp?wnr=5&t=nl&pnr=101&bnr=3), it's sold out.

And I don't expect a reprint. As far as I can tell, mr. Nijssen has moved his interests from database design to the field of knowledge. He now thinks of NIAM as one of the steps he took to arrive where he is now, and he is not interested in pursuing the use of NIAM for database design anymore. He regards his book "Kenniskunde 1a" to be the successor of "Universele Informatiekunde". But though he has added much, he has also left out lots - especially the parts that were primarily useful for those who wwant to use NIAM for database design.

DISCLAIMER: The above is just my personal opinion. The last time I spoke to mr. Nijssen is over 6 years ago - and it wasn't about this subject.

Best, Hugo

-- 

(Remove _NO_ and _SPAM_ to get my e-mail address)