When to cut bait on old IT

Someday you'll find a beloved product/service has fallen down on the job. Don't stick around out of loyalty -- pull the plug

If there's one constant to working in this business, it's that nothing stays the same. On the bright side, new solutions arrive on an almost daily basis -- but not all change is positive. If it hasn't happened to you already, chances are good that someday soon a solution you've depended upon for years will suddenly fail to keep up its end of the bargain.

That failure presents itself in a variety of ways. Sometimes it's a string of buggy software or firmware updates, a raft of inexplicable hardware failures, or a nosedive in the quality of tech support. One of the hardest things in IT is to know when an exisiting solution is past its prime -- and when it's worth trying something new in its place.

[ Also on InfoWorld: Matt Prigge preps you for when the data center goes down and how to get back up. | Sign up for InfoWorld's Data Explosion newsletter for news and updates on how to deal with growing volumes of data in the enterprise. ]

If you've made up your mind to write the Dear John letter, you first must convince the bean counters it's prudent to abandon a solution you've invested in for years. The situation can get sticky if you lobbied heavily to implement that technology in the first place. Of course, this is simply how this industry works -- change arrives, not always for the better -- so avoiding such surprises can be difficult. Fortunately, there are clues that let you recognize the end is coming before you find yourself in a real pinch.

Keep your ear to the ground
One of the most important things you can do is to stay on top of what others are saying about the solutions you're using. That might mean staying active in support forums, even when you're not trying to solve a problem or reading reviews of items you've already bought (especially those that pit your solutions against competitors).

In both cases, knowing what others find good and bad about what you're running will often give you advance warning of trouble. Be on the lookout for negative trends that might indicate issues are afoot, keeping in mind you may also encounter horror stories resulting from a perfect storm of bad circumstances.

Don't be a guinea pig
Keeping your ear to the ground is even more important when a product you rely on undergoes a major upgrade. It's common knowledge that jumping into a brand-new revision or an unproven hardware platform is unwise. However, waiting a few months to upgrade and hoping all bugs have been worked out often isn't good enough, either.

Without a feel for other early adopters' reactions, you may well walk headlong into an unmitigated disaster. (Not to pick on Symantec, but the recently released BackupExec 2012 provides a striking example of this.)

1 2 Page 1
Page 1 of 2
How to choose a low-code development platform