Re: A new proof of the superiority of set oriented approaches: numerical/time serie linear interpolation

Bob Badour wrote:
> Jon Heggland wrote:
>

>> Bob Badour wrote:
>>
>>> Gene Wirchenko wrote:
>>>
>>>> Cimode <cimode_at_hotmail.com> wrote:
>>>>
>>>>> On May 2, 11:18 am, Jon Heggland <jon.heggl..._at_idi.ntnu.no> wrote:
>>>>>
>>>>>> you could say (as D&D do) that a database is a variable---a
>>>>>> dbvar---which has a db value. Which is a tuple.
>>>>
>>>>> Stating that a database is a relation is much safer.
>>>>
>>>>    But likely inaccurate.  It may (and probably will) consist of more
>>>> than one relation.
>>>
>>> If one accepts the idea of relation valued attributes, then one could
>>> look at the dbms as a single tuple with a 0-ary candidate key and an RVA
>>> for each relvar.
>>
>> A single relvar, you mean? Tuples don't have keys. Is a relvar with an
>> empty key safer than a tuple variable? What does "safer" mean?

Cimode's word. I don't know what it means either; that's why I'm asking. I thought you might know, since your post could be interpreted as support for the position that a database is a relation/relvar, as opposed to a tuple variable.

-- 
Jon