Re: teaching relational basics to people, questions

From: paul c <toledobythesea_at_oohay.ac>
Date: Wed, 23 Dec 2009 01:59:15 GMT
Message-ID: <TPeYm.56990$Db2.13981_at_edtnps83>


Bob Badour wrote:
> Jan Hidders wrote:
>

>> On 21 dec, 23:07, r..._at_raampje.lan (Reinier Post) wrote:
>>
>>> Jan Hidders wrote:
>>>
>>>> On 16 nov, 20:42, Sampo Syreeni <de..._at_iki.fi> wrote:
>>>
>>>>> So now I bump into my first real surprise, and the chills immediately
>>>>> go down my spine. That's Date et al.'s answer regarding the
>>>>> implications between 6NF and DK/NF,
>>>>
>>>> athttp://www.dbdebunk.com/page/page/621935.htm
>>>>
>>>>> . In there they flat out state that DK/NF doesn't imply 6NF.
>>>
>>>>> So, my first question is, can this really be true? I mean, this seems
>>>>> highly suspect to me: since 6NF is a normal form like any other
>>>
>>> I don't think so.  It's hard fo me to tell, because I just did the
>>> required math and it turns out don't have $60 to spend on the book
>>> which contains the definition of 6NF required for this discussion,
>>> but if I can get by the Google Books preview, it appears to involve some
>>> degree of interpretation of domain values (as being totally ordered).
>>> Normal normal forms don't do this.
>>
>> Wikipedia is your friend:
>>
>> <http://en.wikipedia.org/wiki/Sixth_normal_form>
>>
>> Whether you can call it a normal form or not is of course largely a
>> matter of definition, but it clearly is different from the other ones.
>> It does for example not remove redundancy or update anomalies, which
>> was pretty much the whole point of the other normal forms.

>
> 1st normal form through 5th normal form involve lossless decomposition
> by project. Fagin proved 5th normal form is necessary and sufficient to
> avoid the update anomalies avoidable by lossless decomposition by project.
>
> It has long been noted that one could create other normal forms using
> different operations. e.g. lossless decomposition by restrict. However,
> to my knowledge, nobody has done so or found a reason to.
>
> If I am not mistaken, 6th normal form does involve lossless
> decomposition by project, which puts it into the same class of normal
> forms as 1st normal form through 5th normal form; even if the reasons
> for choosing 6th normal form do not involve avoiding update anomalies.
>
> Of course, everything is a matter of definition.
>

Both of you succinct as ever. Received on Wed Dec 23 2009 - 02:59:15 CET

Original text of this message