Re: Newbie question about db normalization theory: redundant keys OK?
Date: Sun, 23 Dec 2007 03:05:18 GMT
Message-ID: <Ofkbj.26565$CN4.6827_at_news-server.bigpond.net.au>
David Cressey wrote:
> "Frank Hamersley" <terabitemightbe_at_bigpond.com> wrote in message
> news:Kd6bj.26430$CN4.12955_at_news-server.bigpond.net.au...
>> David Cressey wrote:
[..]
>> Interesting - I had always assumed that Codds use of "time varying" had >> implied temporality but the above quote must weaken that view.
>
> It's my understanding, from reading in here, that "time varying relation"
> was basically the same thing as "relational variable", which in turn gets
> abbreviated to "relvar".
>
> It's my understanding that a consistent view of a database involves
> everything written by prior transactions and nothing written by concurrent
> transactions, let alone future transactions. Concurrency is, of course,
> more complicated than this, but this is sufficient for this discussion.
Consistency for me is taken as a given. I was more looking at the next order problem of temporal datastores which I always assume is exposed to a viewer compliant with all of the Codd RM hallmarks. "Interim" states that could be (practicably) recorded in a temporal datastore must not visible. Otherwise this would be akin to accessing the physical storage engine to second guess the RM - arrgh the horror of that prospect!
I had presumed Codd had addressed temporality in the Principles but now I think not!
[..]
Cheers, Frank. Received on Sun Dec 23 2007 - 04:05:18 CET