1

I have been given a case study of Event Management. I have to develop an entity-relationship diagram (ERD), and based on it check if the resulting relational schema meets the different normal forms. I have developed my ERD based on the SRS that I have developed.

When I converted my ERD to the relational schema I got my tables already normalized.

How can I have nothing left to accomplish with respect to the normalization process after I get an ERD?

Is having nothing left to accomplish after the ERD a desirable situation?

When I am getting my all the relations normalized from the ERD why do I need to do normalization at all?

If we get the database already in 5NF from the ERD then why do we need to think of normalization?

1
  • 2
    Note "normalized" means in 1NF and "fully normalized" means in 5NF or 6NF.
    – onedaywhen
    Commented Oct 19, 2011 at 14:31

1 Answer 1

6

If your tables are already in 5NF then you don't need to think much more about further normalization.

Note however that the principle of full normalization (POFN) is no panacea. There may remain non-obvious update anomalies.

Database constraints will need to be applied. In my experience, designers who start with an ER diagram are great at implementing primary keys and foreign keys but other constraints can be easily overlooked. Other design approaches may result in better constraints e.g. I have a little experience of object role modelling which seemed to work well.

Further normalization should only be considered in conjunction with constraints because 6NF often makes constraints more difficult to implement using today's industrial-strength technologies e.g. none support Standard SQL's CREATE ASSERTION nor CHECK constraints that may include a subquery.

Also consider the principle of orthogonal design (POOD).

1
  • 1
    Object Role Modelling (Halpin) claims it gives 5NF, via involving finding "independent" base variables/components, but it doesn't precisely define what that means, and it ends up being 5NF, so they beg the question. Moreover the obvious everyday notion of "independent" tables is stricter than 5NF. It excludes constraints other than those implied by FDs & JDs & it excludes non-trivial JDs implied by the CKs. So formalizing that notion is needlessly stricter than 5NF. I recall Chen claiming ER gives 3NF but I have no reference.
    – philipxy
    Commented Oct 16, 2023 at 2:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.