Re: Foreign key in Oracle Sql

From: DA Morgan <damorgan_at_x.washington.edu>
Date: Thu, 13 Jan 2005 13:11:29 -0800
Message-ID: <41e6fbb8$1_2@127.0.0.1>

-CELKO- wrote:

>>>On the other side one should acknowledge that not a single major

>
> player in the RDBMS marketplace has actually followed the ANSI
> Committee's recommendations. They take what they want and ignore what
> they don't. <<
>
> Ever hear of FIPS-127 Conformance testing? When that was in place,
> conformance SQL-92 the direction for all of the vendors in ANSI X3H2.
> Then Clinton closed down the program in exchange for a campaign
> contribution from a private testing lab and we started to get drift.
> Mike Gorman did some articles on this that you can find at his
> Whitemarsh Consulting site.
>
> Oracle and IBM pushed the OLAP stuff in SQL-99 together, so that is
> showing up in many other products. The definitions favor contigous
> sequential storage models, but are not procedrually defined as such.
>
> You might want to look at Yukon, Teradata, CA-Ingres, Pervasive, DB2
> and several other smaller players -- all of them do a good job with the
> SQL-92 specs and lots of the SQL-99 features. In fact, SQL is probably
> one of the most portable languages for having so many implementations
> on so many platforms. You can even find SQL dialect translators for
> porting your projects.
>
> Oracle is the worst dialect among the big engines, but it is getting
> better. Maybe they will even have the right data types someday!
> MySQL, and the other "micro-SQL" open source products are really file
> systems with a fancy front end, so I don't count them at all.

I agree completely but to be fair lets keep in mind that ANSI is American like pounds and feet while ISO and much of the other world is metric. I'm not sure Oracle Corp. feels the same degree of need to be "American" as some of the others.

I've worked with Teradata extensively and I wouldn't call its stricter adherence to the ANSI standard a blessing: Rather a curse. That which would have been simple in Oracle was a nightmare. But then that was before they added triggers and proceural options so my feelings may well be prejudiced by that experience.

To me the important is that adding non spec capabilities is not a black mark unless one is forced to use it. So, for example, having the option to use a non-standard datatype is a good thing. Deciding to use it may not be and that is a value judgement.

The standard is effectively dead in that it has almost no influence on vendors and less on buyers. So one can hiss and boo at CONNECT BY but what will win in the end is whether Oracle builds it and developers use it: And we both know they will for a very very long time.

-- 
Daniel A. Morgan
University of Washington
damorgan_at_x.washington.edu
(replace 'x' with 'u' to respond)


----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! >100,000 Newsgroups
---= East/West-Coast Server Farms - Total Privacy via Encryption =---

Received on Thu Jan 13 2005 - 15:11:29 CST