Re: 0.99999998 (was: Unknown SQL)

>I have seen this done (and I had to clear it up!).

I have seen humans commit all sorts of errors. This hardly condemns the tools they used in the commission -- unless of course the tools have significant usability problems that contributed to the error.

>Incidentally, the person who wrote the initial queries was a newcomer
>to SQL with little experience (I notice that people have been claiming
>that SQL is easy to use for newcomers).

You will never hear me make that claim. Of course, you will hear me claim that it is easier to learn than Java for non-programmers.

>This comes down to the debate about the use of unique identifiers.
>- An OODBMS can enforce them (many do implicitly).

Unfortunately, because the uniqueness is not inherent in the data and not under the user's control, an OODBMS will generally assign multiple unique identifiers to a single unique indentity.

>- The relational model can support them but it is up to the user to
>implement them (as you highlight, in most cases the user will do).

Actually, in the relational model, it is up to the DBA or data modeller to implement them. In the OODB situation, it is truly up to the end-user to decide which Bob Badour is really me -- often lacking the necessary information. (Multiple Bob Badour's exist in the real world, and in every case we have distinguishing attributes.)

>However, they are not guaranteed.

It is much likelier guaranteed when the task is under the data modeller's control than when it is strictly under the end-user's control.

Of course, neither network model OODBs nor relational DBMSs have a solution for the end-user who simply ignores the distinguishing characteristics when they are presented. Received on Sun Jul 22 2001 - 01:29:28 CEST