<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
		>
<channel>
	<title>Comments on: SQL generated by 4GLs vs. Oracle stored procedures</title>
	<atom:link href="http://itknowledgeexchange.techtarget.com/itanswers/sql-generated-by-4gls-vs-oracle-stored-procedures/feed/" rel="self" type="application/rss+xml" />
	<link>http://itknowledgeexchange.techtarget.com/itanswers/sql-generated-by-4gls-vs-oracle-stored-procedures/</link>
	<description></description>
	<lastBuildDate>Thu, 20 Jun 2013 11:12:44 +0000</lastBuildDate>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	
	<item>
		<title>By: welcome</title>
		<link>http://itknowledgeexchange.techtarget.com/itanswers/sql-generated-by-4gls-vs-oracle-stored-procedures/#comment-48545</link>
		<dc:creator>welcome</dc:creator>
		<pubDate>Sun, 30 Jan 2005 02:46:10 +0000</pubDate>
		<guid isPermaLink="false">#comment-48545</guid>
		<description><![CDATA[Hi to all,
I answer just for giving 100 points to Ferenc&#039;s answer.
I totally agree with it, in every line and every word !

In my opinion, the so-called &quot;database independent&quot; development leads unevitably to very bad results, at least
in what concerns performance issues.
The &quot;most-common denominator&quot; used for making an application database-independent takes it down to the lowest level, lower than any specific database&#039;s possibilities.
Databases concure to offer better and better results and
facilities, if they all were the same then a single one of
them would suffice for everybody.

It&#039;s a pity not to use Oracle&#039;s wonderful proprietary
additions, in comparison to other databases.
The standard code-generating tools usually do not create
performant code.

At least from my experience, it&#039;s strange to imagine an
organization where the DBA-s SQL knowledge to be wider than
the developers&#039; !
Usually I think rather the opposite is true in most organizations. And this is somewhat natural, considering
that the DBA performs all of the maintenance work, while
the developers are developing both projects and their own
development skills as well.
Every organization should be supposed to have at least a
small group of developers at the highest possible level of
knowledge, even if the DBA team is also at a high level.

I think no code-generating tool could ever compete a
consciuous development done directly by the development
team, based on both the database capabilities and the
business requirements.

My best regards,
Iudith Mentzel

]]></description>
		<content:encoded><![CDATA[<p>Hi to all,<br />
I answer just for giving 100 points to Ferenc&#8217;s answer.<br />
I totally agree with it, in every line and every word !</p>
<p>In my opinion, the so-called &#8220;database independent&#8221; development leads unevitably to very bad results, at least<br />
in what concerns performance issues.<br />
The &#8220;most-common denominator&#8221; used for making an application database-independent takes it down to the lowest level, lower than any specific database&#8217;s possibilities.<br />
Databases concure to offer better and better results and<br />
facilities, if they all were the same then a single one of<br />
them would suffice for everybody.</p>
<p>It&#8217;s a pity not to use Oracle&#8217;s wonderful proprietary<br />
additions, in comparison to other databases.<br />
The standard code-generating tools usually do not create<br />
performant code.</p>
<p>At least from my experience, it&#8217;s strange to imagine an<br />
organization where the DBA-s SQL knowledge to be wider than<br />
the developers&#8217; !<br />
Usually I think rather the opposite is true in most organizations. And this is somewhat natural, considering<br />
that the DBA performs all of the maintenance work, while<br />
the developers are developing both projects and their own<br />
development skills as well.<br />
Every organization should be supposed to have at least a<br />
small group of developers at the highest possible level of<br />
knowledge, even if the DBA team is also at a high level.</p>
<p>I think no code-generating tool could ever compete a<br />
consciuous development done directly by the development<br />
team, based on both the database capabilities and the<br />
business requirements.</p>
<p>My best regards,<br />
Iudith Mentzel</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: ferencmantfeld</title>
		<link>http://itknowledgeexchange.techtarget.com/itanswers/sql-generated-by-4gls-vs-oracle-stored-procedures/#comment-48546</link>
		<dc:creator>ferencmantfeld</dc:creator>
		<pubDate>Wed, 26 Jan 2005 15:39:10 +0000</pubDate>
		<guid isPermaLink="false">#comment-48546</guid>
		<description><![CDATA[concur 100% with reply from geester.

To add, PL/SQL is much more pwoerful than anything Websphere or CA tools can churn out.

Yes, adopt a standard, but don&#039;t get fanatical about it, there are times to make valid exceptions to the rule.

Questions to ask:
Does Websphere have the ability to use the database specific functionality (for instance Oracle Analytic functions, or Oracle&#039;s implementation of TRANSLATE or subquery factoring using WITH clause, using Ref Cursors, custom types, DBMS_% packages, etc.) ? Short answer: in IBM&#039;s dreams !
What about when you need some specific functionality to do something special in Oracle (an insurance company client of mine, wanted a function that accepted an adviser&#039;s id and the function was to return the last appointment for the current month, with date and type, and the first appointment for the next month, this would have involved very complex SQL, but with PL/SQL, the function I wrote executed in 10 milliseconds, and it took me 10 minutes to write and test the function, whereas a SQL centric approach may have been rather expensive, involving many joins and possibole performance concerns and having to worry about the skills of production support staff to maintain the complex SQL once I was off site).

Can WebSpehere generate the SQL required for hierarchical queries (CONNECT BY PRIOR, START WITH, ORDER SIBLINGS BY, NOCYCLE (10g) ?) Same answer again.

Can WebSphere / CA generate Oracle hints when hints are to be used ? Perhaps, but probably not.

When you want a generic SQL generator (like Siebel, PSOFT and SAP do), you are dumbing down the database to the lowest common denominator, which means you are really using the database as a glorified dumb storage bin. The database is there to do powerful and meaningful processing, USE IT !

Getting off my soapbox now. Hope that helped:

Regards:

Ferenc]]></description>
		<content:encoded><![CDATA[<p>concur 100% with reply from geester.</p>
<p>To add, PL/SQL is much more pwoerful than anything Websphere or CA tools can churn out.</p>
<p>Yes, adopt a standard, but don&#8217;t get fanatical about it, there are times to make valid exceptions to the rule.</p>
<p>Questions to ask:<br />
Does Websphere have the ability to use the database specific functionality (for instance Oracle Analytic functions, or Oracle&#8217;s implementation of TRANSLATE or subquery factoring using WITH clause, using Ref Cursors, custom types, DBMS_% packages, etc.) ? Short answer: in IBM&#8217;s dreams !<br />
What about when you need some specific functionality to do something special in Oracle (an insurance company client of mine, wanted a function that accepted an adviser&#8217;s id and the function was to return the last appointment for the current month, with date and type, and the first appointment for the next month, this would have involved very complex SQL, but with PL/SQL, the function I wrote executed in 10 milliseconds, and it took me 10 minutes to write and test the function, whereas a SQL centric approach may have been rather expensive, involving many joins and possibole performance concerns and having to worry about the skills of production support staff to maintain the complex SQL once I was off site).</p>
<p>Can WebSpehere generate the SQL required for hierarchical queries (CONNECT BY PRIOR, START WITH, ORDER SIBLINGS BY, NOCYCLE (10g) ?) Same answer again.</p>
<p>Can WebSphere / CA generate Oracle hints when hints are to be used ? Perhaps, but probably not.</p>
<p>When you want a generic SQL generator (like Siebel, PSOFT and SAP do), you are dumbing down the database to the lowest common denominator, which means you are really using the database as a glorified dumb storage bin. The database is there to do powerful and meaningful processing, USE IT !</p>
<p>Getting off my soapbox now. Hope that helped:</p>
<p>Regards:</p>
<p>Ferenc</p>
]]></content:encoded>
	</item>
</channel>
</rss>

<!-- Performance optimized by W3 Total Cache. Learn more: http://www.w3-edge.com/wordpress-plugins/

Page Caching using memcached
Database Caching 3/10 queries in 0.042 seconds using memcached
Object Caching 281/287 objects using memcached

Served from: itknowledgeexchange.techtarget.com @ 2013-06-20 11:21:49 -->