Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> comp.databases.theory -> Re: Implementing small, fixed domains

Re: Implementing small, fixed domains

From: Hrundi V. Bakshi <Hrundi__V__Bakshi_at_htmail.com>
Date: Mon, 6 Oct 2003 13:12:57 -0700
Message-ID: <k9kgb.22$3Q6.87@news.oracle.com>


"Kenny-Z" <reply_at_group.please> wrote in message news:W_4gb.75$BH5.121691_at_news.uswest.net...
> On one hand, information like this can be modeled as an attribute of
> an entity using integer values (for example), such as 1=Male,
> 2=Female... It is then up to the front-end application to translate
> these into meaningful values. Advantages: less table joins=less
> overhead, simpler to implement the schema.

I really don't see the difference. Logically:

1 = 1*2^0, 2 = 1*2^1+0*2^0
'Male' = 'M'*128^3 + 'a'*128^2+'l'*128^1+'e'*128^0 where 128 is ascii range.

Physically, any reasonable database implementation should be able to compress a column storing {'Male','Female'} to something close to 1 bit/per row. Received on Mon Oct 06 2003 - 15:12:57 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US