Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Mailing Lists -> Oracle-L -> Re[2]: Normalization
On normalisation - it is probably worth pointing out that - although
eliminating redundancy is indeed an important part of the process - it
is only a part. Other activities include resolving inter-key and
inter-attribute dependencies.
Irrespective of the particular mechanisms applied in the process of normalisation (of which the above are examples), the whole thrust is to ensure that one creates a logical, internally consistent data model, which is then implemented by the RDBMS - and that is the great strength of the RDBMS - it supports the physical implementation of a correctly designed logical data model (well, with certain caveats according to the product).
The process of normalisation is the product of sound reasoning. The benefit of applying the process is that, if the results of a data query should ever be challenged, one can turn to the data model design and demonstrate totally unambiguously that there are no inconsistencies in that design, therefore the query outputs MUST be correct. Err, this could be rather important...
If you are querying from a non-normalised data model, you can never claim that assurance. And if that assurance is not required, well, as Jared said, why bother with the expense of an RDBMS - use a text file with sed/grep etc....
peter
edinburgh
--
mailto:pgro_at_bgs.ac.uk
BGS. . http://www.bgs.ac.uk *********************************************************************Please see the official ORACLE-L FAQ: http://www.orafaq.com
----------------------------------------------------------------