Re: codd's 12 Rules

Well, since you're saying it's not your opinion, Vern, then I'm not
stepping on your toes by saying that this is, in my opinion (one which I
won't even attempt to hide behind false humility), one of the most
ridiculous assertions ever made in all of programming. To boil it down,
even if the DB2 SQE follows every other stricture and supports every
other feature of SQL (which it does), then it still wouldn't be
relational because you can ALSO access the data via RLA.

Well, to extend that incredibly myopic point of view, then EVERY
database system is non-relational because at the end of the day you can
use a hex editor to change the data in the file.

So, to anyone who says that RLA invalidates the relational nature of
DB2, I say go Hex yourself.

:)

I know - go do a google search yourself, as I did - look at the various interpretations - you will see this narrow view. Not mine - just reporting. You didn't quote the bits where I said just this - that there ARE different points of view.

But this fact - that we can use DDS to generate and modify tables - this is held up by many to say that this is NOT a relational system. My friend, Dave Odom, is one who tried to tell us this. I don't always agree with him, of course.

Vern

----- Original Message -----

Rule 0 may already be broken

Rule 0: The system must qualify as relational, as a database, and as a management system.


Does IBM i DB not qualify as relational? As a database? As a management system?

since there is a non-relational way to manage the database.

Isn't that a rather narrow interpretation of the rule? And a narrow interpretation of a management system?

-Nathan