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

Home -> Community -> Usenet -> c.d.o.server -> Re: MySQL 4

Re: MySQL 4

From: <ctcgag_at_hotmail.com>
Date: 09 Aug 2003 21:50:43 GMT
Message-ID: <20030809175043.861$AH@newsreader.com>


Daniel Morgan <damorgan_at_exxesolutions.com> wrote:
> ctcgag_at_hotmail.com wrote:
>
> > <snipped>
> >
> >
> >
> > So when I pull a future date out of my ass for the DOB, it yells at me
> > and tells me to reach in for a past date for the DOB? That's a
> > functionality I think I can do without, if the price is right.
> >
> > Xho
> >
> > --
> > -------------------- http://NewsReader.Com/ --------------------
> > Usenet Newsgroup Service New Rate! $9.95/Month 50GB
>
> Then obviously your experience with medical, insurance, and other
> demographic applications is rather limited.

Yes, it is rather limited.

> I can count in the thousands the times I have had to clean up databases
> where people were born and transactions occured in the future ...

Why did you have to clean them up? Or, how did you clean them up? Did you call each person and ask them for a corrected date?

It seems to me that if you store the date, you want the date to be correct. If you only want it to be reasonable but not necessarily correct, then you probably don't need to store it at all. If checking for dates in the future eliminated >95% of date errors, then it would seem useful to me. But if it only eliminates a (relatively) few obvious errors, I wouldn't hold my breath for that feature.

> sometimes hundreds of years in the future. The City of Seattle Washington
> has a database of which I am familiar in which many things are entered in
> the years 2991 and 3001 and that is hardly an exception to the rule.

Maybe that's their expected date to pay of their debt.

If that many gross errors are occuring, I'd assume there are even more errors which are more subtle. I'd rather track down the root cause of these errors, rather than to implement methods that reject only the most recognizable errors, thus disguising the true error-rate.

Not that I could'd find some value in the ability to do things like this, but it's definitely way down on my wish list for added MySQL functionality.

But I'm certainly not a insurer or demographer.

Xho

-- 
-------------------- http://NewsReader.Com/ --------------------
Usenet Newsgroup Service              New Rate! $9.95/Month 50GB
Received on Sat Aug 09 2003 - 16:50:43 CDT

Original text of this message

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