The Pythian Group HomeEmergencyClient LoginContact Us
About Us   Our Services   The Pythian Advantage   News
back

News

Pythian Papers

Finding Your Place on the Technology Curve:
Avoiding Oracle's bleeding edge.

by Michael S. Abbey

A passion for excellence  

The cutting edge. Most of us in the IT Industry have reveled in its advantages at some point. Unfortunately, too many of us have also witnessed the resulting carnage when the cutting edge becomes the bleeding edge. All technology has its bleeding edge and Oracle is no exception. Sophisticated, ever-evolving software like Oracle has a complicated development cycle, which makes it all the more difficult to determine the right point to jump in.

Like most technology companies, Oracle has a vested interest in getting their customers to upgrade early and often. Why? Two main reasons:

  1. The quicker we upgrade, the quicker Oracle can phase out older buggier versions. This makes strategic sense, especially since Oracle also acts as a support organization.
  2. Oracle needs their new technology test-driven in the marketplace. Internal testing just isn't the same as putting it out in the real world. Our early adoption of their latest technology gives Oracle the opportunity to fine-tune their product for future releases.

Of course, what's best for Oracle isn't always what's best for your company. When you're trying to decide if it's time to upgrade, here are some things to consider:

  • Often third party applications are not certified with the latest Oracle technology. The mingling of other vendors' software with Oracle's needs to influence your upgrade decisions.
  • Many companies only use the RDBMS Server, which is the backbone of Oracle's technology, but a very small player in the vast suite of products rolled together in Oracle version XYZi. If you are only planning to upgrade the RDBMS Server, early adoption is substantially more tempting and feasible.
  • It's important to evaluate realistically how much time your personnel has to test-drive your applications against the new release. Don't plan in theory—make sure you've reserved the hours in the schedule.

You've taken stock and think you want to progress down the aggressive upgrade path? Here are a few things to keep in mind:

  • Proceed with extreme caution on any initial major release of Oracle software. These major releases have not undergone any real marketplace testing and are therefore prone to bugs that haven't yet been identified. Subsequent versions are usually fixes of the bugs discovered by early adopters.
  • The bugs that have been fixed in the new release are, without faulting anyone in particular, not always as "fixed" as we are led to believe. While we enjoy the fixes that are present, it's still possible to get bitten by the old bugs—as well as new ones.
  • Oracle's support organization is not as skilled at supporting new releases from day one. Yes, their analysts undergo extensive training in conjunction with any new release, but they simply aren't as seasoned as they were with previous offerings and are sometimes learning while assisting you.

Still trying to make up your mind? When you're unsure what your team and/or system can handle, it's usually best to err on the side of caution. Here's an easy guideline to follow: Hold off on adopting any brand new major releases, but once you've got them, move full-speed-ahead with any bug-fixing upgrades. The cutting edge can become the bleeding edge all too quickly, but with careful planning you don't have to be the one to fall on the sword.

Want more advice on upgrading Oracle? See our checklist