Re: Dreaming About Redesigning SQL

"Steve Lancour" <stevel_at_lancour.com> wrote in message news:N9WdnZe9vtmD4D-iRVn-iw_at_comcast.com...
> Bob Badour wrote:
> > "Steve Lancour" <stevel_at_lancour.com> wrote in message
> > news:CsGdnZoN95XJJjyiRVn-jw_at_comcast.com...
> >
> >>Mikito Harakiri wrote:
> >>
> >>
> >>>"Steve Lancour" <stevel_at_lancour.com> wrote in message
> >>>news:mKudnYI38-U5KTyiRVn-gw_at_comcast.com...
> >>>
> >>>
> >>>>No, I postulate only one mistake. If the DBA makes a mistake in
setting
> >>>>up the constraint the applications may be perfect and yet still
corrupt
> >>>>data. Same as if there were a mistake in the example Pick code I
> >
> > posted.
> >
> >>>
> >>>A mistake in high abstraction language is much less likely then in low
> >
> > level
> >
> >>>language.
> >>>
> >>>
> >>
> >>I don't know. Perhaps. But if we can agree that the QA efforts on both
> >
> >
> > We cannot agree on that point. There are too many database operations
that
> > never go through QA.

>
> Perhaps in systems you've built but not in mine.

Really? You have never had a user ask you to investigate suspicious data? Your boss has never asked you to deliver a complicated summary on short notice for a big meeting?

You must have patient users. Received on Sat Nov 01 2003 - 00:16:40 CST