Re: The Practical Benefits of the Relational Model

In article <ap0iuv$11i0$1_at_sp15at20.hursley.ibm.com>, Paul Vernon <paul.vernon_at_ukk.ibmm.comm> wrote:
>
>Database constraints (often with a temporal component) can be used by users
>(when allowed by the database design) to create guarantees about whether
>values just read are still valid. Such a constraint based approach not only
>exposes all 'locks' as relational data, but also allows different relvars
>(and sets of relvars) within a database to have different locking
>'protocols' implemented. It is my position that locking is a database
>design issue rather than a DBMS 'feature'.

I agree, but to me that means that the DBMS should allow the DBA to choose the appropriate locking protocol. However, your suggestion seems to be to let all the applications do their own locking. I cannot imagine that you really think that this will be easier to understand for the application builders then the concept of transaction.

By the way, using predicates/constraints for locks is done in the socalled 'predicate locking protocol'. Would that be similar to what you are proposing? It's not the easiest of protocols.

If you want to know more about locking and transaction management I can recommend the book "Transactional Information Systems" by Gerhard Weikum and Gottfried Vossen:

  http://www.mkp.com/books_catalog/catalog.asp?ISBN=1-55860-508-8#Contents

I believe they talk in section 8.2 about on predicate locking.

• Jan Hidders