Re: Proposal: 6NF
Date: Wed, 04 Oct 2006 01:20:17 GMT
Message-ID: <l%DUg.87731$5R2.64514_at_pd7urf3no>
David Cressey wrote:
> "paul c" <toledobythesea_at_oohay.ac> wrote in message
> news:8dSTg.78337$R63.70243_at_pd7urf1no...
>> David Cressey wrote: >>> "Cimode" <cimode_at_hotmail.com> wrote in message >>> news:1159707552.227184.276470_at_k70g2000cwa.googlegroups.com... >>>> .. >>> What's your proposal for a systematic way of dealing with missing data? >>> >>> >> David, PMFJI, I really do think that an enterprise that is truly >> depending on missing data doesn't have a problem that is limited to its >> db, as it will soon go out of business. Actually, I don't really see >> how one can depend on missing data in general and no systematic solution >> is called for. Heh, also think that the IBM approach is best - namely >> It's Better Manually! >>
>
> Paul,
>
> My phrase "a systematic way of dealing with missing data" is from Codd's
> twelve rules. I think Codd got this one right. Not because the RDM is any
> more dependent on dealing with missing data than any other data model might
> be, but because, in the real world, you are going to be faced with the
> reality that data that "ought to be there" isn't there.
>
> You either have a systematic way of dealing with missing data, or you deal
> with missing data in an unsystematic way, when the eventuality happens.
> Which is worse?
>
>
> I didn't intend the inference that you drew, that a company should design
> its data strategy around dependency on missing data. I do think that a
> system that a company depends on should not choke up and refuse to work if
> some item of expected data is not present.
>
> I also don't believe that "it's better manually". Manual systems also have
> ways of dealing with missing data. Sometimes they depend on "common sense"
> on the part of the people in the system. Sometimes that works. Sometimes
> it doesn't. I think a good, well designed system that incorporates a
> systematic way of dealing with missing data can be significantly more robust
> that a system that relies on uncommon levels of common sense.
>
>
>
>
>
David, regarding the "inference", I wasn't trying to put words in your mouth, just making what might seem an extreme statement to some in order to make my point. As for the "it's better manually" comment, I meant it only as a joke, although I have noticed from time to time that certain manual systems were far more efficient and more "cost-effective" than the computerized systems that replaced them!
I didn't mean to distract from the technical topic. I imagine I'm one of the few people who likes Hugh Darwen's approach, even though I admit it can seem ponderous (ie., many, many tables) to some programmers. Then there is Fabian Pascal's approach which is slightly less pragmatic since it depends on an implementation that doesn't exist as far as I know, but which seems also a little more ideal to me. My attitude about the systems I've seen is that they could do with some ruthless expunging of table columns that were added in haste. Whenever I was able to get to the REAL user or the big boss, it was surprising how many requirements cooked up by consultants and ordinary users turned out to be not essential, just subjective "nice-to-haves".
magoo Received on Wed Oct 04 2006 - 03:20:17 CEST