Re: First Impressions on Using Alphora's Dataphor

From: Mikito Harakiri <mikharakiri_at_iahu.com>
Date: Fri, 27 Aug 2004 10:03:34 -0700
Message-ID: <D0KXc.47$zw4.47_at_news.oracle.com>

"Leandro Guimaraens Faria Corsetti Dutra" <leandro_at_dutra.fastmail.fm> wrote in message news:pan.2004.08.27.12.31.59.829172_at_dutra.fastmail.fm...
> Em Thu, 26 Aug 2004 10:10:56 -0700, Josh Hewitt escreveu:
>
> > Is it not better then to forget about UDTs in the database and only
> > allow them in programming languages where developers have complete
> > control over them?
>
> Why should developers have free reign over UDTs? There are
> sure important enough to be centrally managed by DBAs, and specified
> by system analysts.
>
> If the company is too bureaucratic and thus managing UDTs
> becomes a problem, that's an organisational problem, not a
> technological one.

Leandro, do you really believe that couple of thin paragraphs is enough to rebuff insightful and quite substantial original message?

To add to UDT problems (which were posted sometime ago in the c.d.t exchange with Paul Brown, so I plot only summary here):

1. Creating User Defined indexes is hard.
2. User Defined statistics is expert users domain.
3. UDT optimization is virtually non-existent.

Even though 1-3 might be not fundamental problems, they create unsatisfactory Return-On-Investment when creating a UDT of only limited utility requires unproportional effort. Received on Fri Aug 27 2004 - 19:03:34 CEST