paul c wrote:
> Bob Badour wrote:
>
>> paul c wrote:
>>
>>> Bob Badour wrote:
>>> ...
>>>
>>>> Any two values of differing most-specific-types are unequal. Thus,
>>>> if x and y each have an attribute z where the types are different,
>>>> the join will have cardinality zero and will have an attribute z
>>>> with the most specific supertype of the z attributes in x and y.
>>>> This might be the universal supertype.
>>>> ...
>>>
>>> Thanks very much for that. The word 'if' in 'if x and y each have an
>>> attribute ... ' might be at the centre of my question. Will ponder
>>> that, it might be the first time that anybody has tried to tell me in
>>> terms that I could dig why some OO notions might be important.
>>
>> Which OO notions would those be?
>
> Inheritance and inheritance.
Sub-typing bears little resemblance to inheritance as far as I can tell.
Received on Mon Jul 03 2006 - 07:51:06 CDT
