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: Query execution for intervals

Re: Query execution for intervals

From: Tony Rogerson <tonyrogerson_at_sqlserverfaq.com>
Date: Tue, 7 Feb 2006 08:31:24 -0000
Message-ID: <ds9ls5$v3$1$8300dec7@news.demon.co.uk>


Hi Frank,

Nope, the point is that a column defined NULLable cannot be part of a primary key.

My real point is that celko makes fundemental mistakes all the time, yet, he bashes newbies who do the same - so, who bashes him? I think the more experienced among us need to pick him up on things and perhaps this arrogant, condesending and ignorant attitude he puts across to newbies might change.

> (Most definitely and by deliberate choice not an MVP - just my own
> person)!

I don't work to get the MVP, I just love helping people in the SQL community and especially here in the UK; Microsoft seem to want to keep giving me the award which has a lot of advantages ;)

-- 
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials


"Frank Hamersley" <terabitemightbe_at_bigpond.com> wrote in message 
news:ryQFf.245264$V7.6879_at_news-server.bigpond.net.au...

> Tony Rogerson wrote:
>> So why say this: >> >> >> try putting the end_date into the >> >> PRIMARY KEY to get a covering index >> >> Its about time you started checking your answers, perhaps if you spent a >> lot less time trying to be clever by putting posters down you might not >> make so many fundemental mistakes. >
> Tony my man - how long have you been waiting to stick one on CELKO?
>
> If so then IMO you have blown it. I clearly understood both at a semantic
> and physical level the intent of his post, and was almost going to post
> similarly except I was beaten to the punch! Just because end_date allows
> nulls this does not invalidate the technique proposed for the OP's
> interest.
>
> I could be wrong but as I read it you seem to be labouring under the
> misunderstanding that a PK must be a single attribute in a relation.
>
> Cheers, Frank.
> (Most definitely and by deliberate choice not an MVP - just my own
> person)!
Received on Tue Feb 07 2006 - 02:31:24 CST

Original text of this message

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