Re: I have read tons of theory...but still...one question

"Kostas" <noemail_at_noemail.net> wrote in message news:10n04s69tvgu11b_at_corp.supernews.com...

> >
> > In the case of cellulars, a phone knows it's "contract number" in
> > America.
> > In Europe, think there's a card inside to phone that knows its contract
> > number. The contract number determines the phone number.
> >
>
> Alright, but this goes a little bit too far for my purposes. A phone
> essentially means a number at which a person can be reached.

OK, it's your choice. But I remember a few years back we had a database that was being used for the "roaming charge clearinghouse" process in the US. When it was proposed to extend this software to cover European cellulars as well, the database was incomplete, for precisely this reason! Sorry, but I don't remember the particulars, and maybe it isn't true any more.

> Yes, physical redundancy and also the case of mispellings. If I store the
> phone types once and correctly I dont need to worry whether there will be
a
> mispelling in entry of one fact such as "CelPhone" instead of "CellPhone"
or
> "Cell_Phone".

I'd like to strongly suggest that mispellings are a user interface issue and NOT a database issue. I think there's value in keeping those two groups of issues separate. That's not to say your choice is the wrong one, but it ain't database theory that's going to address that.

BTW, there are ways to design the UI so that its dropdown list is built from all the valid (and correctly spelled) values.

>
> > If that's what you are thinking, then you are vulnerable to what I'll
> > call
> > a "Neo Con" (being conned by Neo).
>
> The only Neo I know is the one from Matrix :-)

There's another Neo that comes around here every once in a while. Every time someone wins his contest, he changes the definition of "redundancy". Received on Fri Oct 15 2004 - 20:38:02 CEST