Re: Little question for RDM theoristes

> So you agree that changing the header of a certain relation variable
> *requires* one to do something about the current value of that
> variable, i.e. it *must* be assigned *another* value than the one that
> is currently there ?
>
> How can one explain this to be a *requirement* if it does not derive
> from the fact that R1 has changed type (and that the current value
> simply isn't of that new type) ?

Look on this two cases
1)After adding new _tuple(s)_ into variables we get new value which is result of UNION of two relvalues (first one is previous value,second one is set of added tuples).
2)After addind new _attribute_ into variables we get new value too which is result of carthesian product of two relvalues (first one is previous value,second one is unary relation).

There is no difference. In both cases we get new values using peration of relational algebra and in both cases these values is ones of the same and only type. Received on Fri Jun 16 2006 - 12:21:25 CEST