As Published In
Oracle Magazine
September/October 2009

From Our Readers


Your corrections, your opinions, and your requests: Here’s your forum for telling us what’s right and wrong in each issue of Oracle Magazine, and for letting us know what you want to read.

Now We’re Cooking

Oracle has released a version of Oracle Database 10g for the Mac OS X platform—finally. This is great news. As a 20-year user of SunOS when I was working, I can now enjoy Oracle tools on my iMac in my retirement years.

Great job to all those involved in the development and in providing a cookbook-style install.
Michael Morin
mmorin@cassblue.com

The editors reply: For more information about Oracle Database on the Mac, visit the Oracle-on-Mac OS X Technology Center.

There you’ll be able to find the Oracle Database 10g Release 2 (10.2.0.4.0) download and documentation, as well as product reviews, install guides, and blogs.

More Pages for Coding

I really like Oracle Magazine . I am a high school student in Indonesia, and I can never wait until the next issue comes out. I would love to see you provide special pages in the magazine—or on your Web site—to demonstrate connecting Oracle technology with Visual Basic, .NET, C, PHP, and other languages.
Sinclair Pahlevi
adam.pahlevi@gmail.com

Two Ps in a Pod

Two questions came to mind in reading Tom Kyte’s latest column, “On Popularity and Natural Selection” (Oracle Magazine, July/August 2009).

First, in the statement “P1 would be invalid . . . ,” don’t you mean P2? P1 has just been recompiled. P2 is dependent on P1.

Second, in the statement “Oracle Database 10g added . . . ,” do you mean Oracle Database 11g? The context of this discussion is the new compilation enhancements introduced in Oracle Database 11g.

I hope I’m not mistaken in these assumptions.
Ted Persky
tpersky@bellsouth.net

Tom Kyte replies: First, thanks! Second, you are correct. The following sentence

. . . Now, in Oracle Database 10g and before, P1 would be invalid, but because we were using Oracle Database 11g and we did not change the signature of the procedure, we did not invalidate the dependent code. . . .

should read

. . . Now, in Oracle Database 10g and before, P2 would be invalid, but because we were using Oracle Database 11g and we did not change the signature of the procedure, we did not invalidate the dependent code. . . . 

E-Mail the Editors


Send your opinions about what you read in Oracle Magazine, and suggestions for possible technical articles, to opubedit_us@oracle.com. . Or click the Write the Editors link on our Website, oracle.com/oraclemagazine. You can also follow our @oraclemagazine Twitter feed or join us on Facebook. Letters may be edited for length and clarity and may be published in any medium. We consider any communications we receive publishable.
As to your second question about the reference to Oracle Database 10g , that was intentional. It was Oracle Database 10g that added the “check to see if we need to recompile first” logic. I could definitely have made that more clear—sorry about that.

The editors add: The correction is online.

More to the Point

“Restore to the Point” by Arup Nanda (Oracle Magazine, November/December 2006) is a great article. I’ve used it to implement flashback and restore points on our test databases.

I’d be interested to see details on what happens to the files used for the restore points after a restore point has been dropped. Perhaps that’s in the next article.

Tim Chidlow
tim.chidlow@eon-uk.com



Send us your comments