Comments for Crazy Idea http://anjanb.wordpress.com Experiences of a geek-to-be Sun, 20 Jul 2008 07:58:47 +0000 http://wordpress.org/?v=MU Comment on RFE(Req For Enhancement) java.net : Add SVN by LUfjOuon http://anjanb.wordpress.com/2006/03/04/rfereq-for-enhancement-javanet-add-svn/#comment-3567 LUfjOuon Wed, 18 Jun 2008 10:31:37 +0000 http://anjanb.wordpress.com/2006/03/04/rfereq-for-enhancement-javanet-add-svn/#comment-3567 best <a href="http://darzas.za.pl" rel="nofollow">best</a> best
best

]]> Comment on Is anybody in production using Java 6 by anjanb http://anjanb.wordpress.com/2007/06/09/is-anybody-in-production-using-java-6/#comment-3533 anjanb Thu, 20 Mar 2008 20:56:49 +0000 http://anjanb.wordpress.com/2007/06/09/is-anybody-in-production-using-java-6/#comment-3533 I heard that yahoo's search team is in production using java 6 along with Hadoop, etc. How about other guys doing enterprise stuff ? I heard that yahoo’s search team is in production using java 6 along with Hadoop, etc.

How about other guys doing enterprise stuff ?

]]> Comment on Java 6(Mustang) upgrade path hurdles ? by dorette http://anjanb.wordpress.com/2006/12/21/java-6-upgrade-path-hurdles/#comment-3081 dorette Mon, 22 Oct 2007 15:19:11 +0000 http://anjanb.wordpress.com/2006/12/21/java-6-upgrade-path-hurdles/#comment-3081 http://travelnursingdirect.com/ http://travelnursingdirect.com/

]]> Comment on Jdbc : Why doesn’t Oracle’s driver support READ_UNCOMMITTED ? by Dr. Java http://anjanb.wordpress.com/2007/09/07/jdbc-why-doesnt-oracles-driver-support-read_uncommitted/#comment-2990 Dr. Java Wed, 03 Oct 2007 11:33:49 +0000 http://anjanb.wordpress.com/2007/09/07/jdbc-why-doesnt-oracles-driver-support-read_uncommitted/#comment-2990 For your convenience, Tom's answer can be found <a href="http://www.oracle.com/technology/oramag/oracle/05-nov/o65asktom.html" rel="nofollow">on his site.</a> For your convenience, Tom’s answer can be found on his site.

]]> Comment on Jdbc : Why doesn’t Oracle’s driver support READ_UNCOMMITTED ? by peter veentjer http://anjanb.wordpress.com/2007/09/07/jdbc-why-doesnt-oracles-driver-support-read_uncommitted/#comment-2781 peter veentjer Sat, 08 Sep 2007 15:19:34 +0000 http://anjanb.wordpress.com/2007/09/07/jdbc-why-doesnt-oracles-driver-support-read_uncommitted/#comment-2781 Databases that use Multiversion concurrency control (oracle, postgresql, mysql&innodb, firebird etc) don't need dirty reads. A dirty read is useful if you want to do a non blocking read, but Multiversion Concurrency Control (MVCC) database have a different solution. Instead of reading uncommitted data, a previous committed version of a record is used (hence the name multiversion). This approach is much saver because you can't read uncommitted data (so data that never made it into the db) or data that is not valid (data could be written before its constraints are checked). That is why a READ_UNCOMMITTED isolation level is ugraded to READ_COMMITTED. A similar explanation can be used for REPEATABLE_READS. The SERIALIZED isolation level also prevents unrepeatable reads from happening and the serialized isolation level is quite easily realized in oracle (also based on the multiversion mechanism and optimistic locking at the end of serialized transaction commit). That is why REPEATABLE_READ is automatically upgraded to SERIALIZED. Databases that use Multiversion concurrency control (oracle, postgresql, mysql&innodb, firebird etc) don’t need dirty reads. A dirty read is useful if you want to do a non blocking read, but Multiversion Concurrency Control (MVCC) database have a different solution. Instead of reading uncommitted data, a previous committed version of a record is used (hence the name multiversion). This approach is much saver because you can’t read uncommitted data (so data that never made it into the db) or data that is not valid (data could be written before its constraints are checked).

That is why a READ_UNCOMMITTED isolation level is ugraded to READ_COMMITTED.

A similar explanation can be used for REPEATABLE_READS. The SERIALIZED isolation level also prevents unrepeatable reads from happening and the serialized isolation level is quite easily realized in oracle (also based on the multiversion mechanism and optimistic locking at the end of serialized transaction commit). That is why REPEATABLE_READ is automatically upgraded to SERIALIZED.

]]> Comment on Jdbc : Why doesn’t Oracle’s driver support READ_UNCOMMITTED ? by anjanb http://anjanb.wordpress.com/2007/09/07/jdbc-why-doesnt-oracles-driver-support-read_uncommitted/#comment-2772 anjanb Sat, 08 Sep 2007 00:48:01 +0000 http://anjanb.wordpress.com/2007/09/07/jdbc-why-doesnt-oracles-driver-support-read_uncommitted/#comment-2772 Hi Morgan, Thank you. Guess it's time to refresh my Oracle fundas by reading the WROX PRESS "EXPERT ONE-ON-ONE ORACLE". I'm sure Tom would give me a hint as to why they do it. Do, do other DBs(like MS-SQL, MySQL, DB2) allow Dirty Reads ? How about Java DB, HSQL, etc ? Thanks again, BR, ~A Hi Morgan,

Thank you.

Guess it’s time to refresh my Oracle fundas by reading the WROX PRESS “EXPERT ONE-ON-ONE ORACLE”. I’m sure Tom would give me a hint as to why they do it.

Do, do other DBs(like MS-SQL, MySQL, DB2) allow Dirty Reads ? How about Java DB, HSQL, etc ?

Thanks again,

BR,
~A

]]> Comment on Jdbc : Why doesn’t Oracle’s driver support READ_UNCOMMITTED ? by Morgan http://anjanb.wordpress.com/2007/09/07/jdbc-why-doesnt-oracles-driver-support-read_uncommitted/#comment-2770 Morgan Fri, 07 Sep 2007 21:34:18 +0000 http://anjanb.wordpress.com/2007/09/07/jdbc-why-doesnt-oracles-driver-support-read_uncommitted/#comment-2770 Because transaction isolation levels READ_UNCOMMITTED and REPEATABLE_READ don't exist in Oracle databases by design. Because transaction isolation levels READ_UNCOMMITTED and REPEATABLE_READ don’t exist in Oracle databases by design.

]]> Comment on openJDK/Java 7 : RFE by David Herron http://anjanb.wordpress.com/2007/08/30/openjdkjava-7-rfe/#comment-2652 David Herron Fri, 31 Aug 2007 00:02:26 +0000 http://anjanb.wordpress.com/2007/08/30/openjdkjava-7-rfe/#comment-2652 You should bring this suggestion to one of <a href="http://mail.openjdk.java.net/" rel="nofollow">the OpenJDK mailing lists</a>, look at the <a href="http://openjdk.java.net/contribute/" rel="nofollow">contribution process</a> to see the outline of how to proceed. This probably wouldn't require a full JSR thingy. There are various management interfaces like <a href="http://java.sun.com/javase/6/docs/technotes/guides/attach/index.html" rel="nofollow">Attach API</a>, <a href="http://java.sun.com/javase/6/docs/technotes/guides/jvmti/index.html" rel="nofollow">JVMTI</a>, and <a href="http://java.sun.com/javase/6/docs/technotes/guides/management/index.html" rel="nofollow">Monitoring and Management</a> which provide similar features. You should bring this suggestion to one of the OpenJDK mailing lists, look at the contribution process to see the outline of how to proceed. This probably wouldn’t require a full JSR thingy.

There are various management interfaces like Attach API, JVMTI, and Monitoring and Management which provide similar features.

]]> Comment on Java Swing Vs VB 4.0 by anjanb http://anjanb.wordpress.com/2007/07/17/java-swing-vs-vb-40/#comment-2009 anjanb Wed, 18 Jul 2007 19:33:47 +0000 http://anjanb.wordpress.com/2007/07/17/java-swing-vs-vb-40/#comment-2009 Hi Arun, Thanks for stopping by. After posting it, I briefly thought that someone could misunderstand where I was coming from. Let me clarify. What I mean is this "Although Java and Swing have improved a lot in the last few years to be compelling considerations for desktop apps, developer support for building such GUIs was not quite there. With Netbeans/Matisse and a few open source libraries/toolkits/frameworks, Java can be a good tool to develop such apps. I know that Sun employees recently came out with a new book called "Filthy Rich Java apps" or something like that. So, I know that one can build such apps TODAY, thousands of VB developers have been able to do that since the last 10 years. I emphasis was ON developer tools NOT the runtime capabilities of the platform. BR, ~A Hi Arun,

Thanks for stopping by.

After posting it, I briefly thought that someone could misunderstand where I was coming from.

Let me clarify. What I mean is this “Although Java and Swing have improved a lot in the last few years to be compelling considerations for desktop apps, developer support for building such GUIs was not quite there. With Netbeans/Matisse and a few open source libraries/toolkits/frameworks, Java can be a good tool to develop such apps. I know that Sun employees recently came out with a new book called “Filthy Rich Java apps” or something like that.

So, I know that one can build such apps TODAY, thousands of VB developers have been able to do that since the last 10 years. I emphasis was ON developer tools NOT the runtime capabilities of the platform.

BR,
~A

]]> Comment on Java Swing Vs VB 4.0 by Arun Kumar http://anjanb.wordpress.com/2007/07/17/java-swing-vs-vb-40/#comment-1992 Arun Kumar Wed, 18 Jul 2007 01:16:37 +0000 http://anjanb.wordpress.com/2007/07/17/java-swing-vs-vb-40/#comment-1992 I would have to disagree entirely with your opening line ("Java’s weak point has been its Rich client programming capabilities")! Check out Kerika (www.kerika.com), which is written entirely in Java and provides a very rich client indeed. And check out http://www.kerika.com/2007/06/why-we-chose-java.html on why we chose to build Kerika in Java. I would have to disagree entirely with your opening line (”Java’s weak point has been its Rich client programming capabilities”)! Check out Kerika (www.kerika.com), which is written entirely in Java and provides a very rich client indeed. And check out http://www.kerika.com/2007/06/why-we-chose-java.html on why we chose to build Kerika in Java.

]]>