A disaster is waiting to happen

What happened to California? How, in a single generation, did it go from the land where it never rains to the land of rolling blackouts and the dot-com dust bowl? Not to mention earthquakes, mudslides, brush fires …

Which, as an IT guy, makes me think of disaster recovery.

Ten years ago, I’d be asking if your organization had a disaster recovery plan. Today, I’m pretty sure I don’t have to ask. Everyone recognizes that planning for the worst is just prudent.

But have you looked at your plan lately? Some organizations simply create a plan, purchase a service, check it off on their security audit reports, then go about their business as usual — and most of the time, that’s just fine. Until disaster strikes.

So what constitutes a disaster? It could be as serious as a hurricane, an earthquake, or a flood — or as common as an unplanned power outage. You need to be prepared for various contingencies, because your response will vary according to the nature of the problem. For example, you need to get backup power running as soon as possible after a blackout, but going live immediately after a flood could cause more damage than the disaster itself.

I’m not going to document the process of creating a business contingency plan here. There are plenty of resources available on the Web, some of which I’ve listed below, and scads of companies that provide disaster planning services. I suggest reviewing some disaster planning articles every year just to keep the concept fresh in your mind.

It’s also good policy to hold an annual disaster drill. Vary the emergency, but make sure your crucial people know what to do. Chances are, you’ll uncover unforeseen flaws in your plan — an empty fuel tank for your emergency power generator, a new phone number for a key staffer, someone forgetting to document the new password at your off-site storage facility.

Several organizations offer training and certification programs for both business contingency planning and full-scale disaster recovery. A degree program in disaster recovery is overkill for any organization that doesn’t actually provide disaster recovery services, but it might make sense to send a staffer through some certification courses.

If you have a plan in place and you’re sure your employees are prepared to execute it if necessary, relax. Chances are you’ll never have to use the plan. Don’t keep looking over your shoulder for a 747 to fall on your data center. You have plenty of other things to worry about.

Good general intros:

• Safety Central: Business contigency planning
• How to plan for enterprise-wide business and service continuity

Organizations:

• DRI International
  (originally the Disaster Recovery Institute)

Web sites:

• 
  Disaster Resource Guide

• Survive: The Business Continuity Group

Events:

• Corporate Contingency Planning Seminar and Exhibition

The arrogance of the OS majority

In my last column, I noted that when overzealous proponents of some fringe operating systems rant about criticism of their favorite platforms, they hinder rational discussion. But supporters of mainstream platforms suffer from a different problem: arrogance and complacency.
Microsoft’s operating systems run the majority of corporate desktops (and those of home users, for that matter). How did this happen? By a combination of useful software, savvy marketing, and predatory licensing practices that virtually forced computer hardware vendors to bundle Microsoft operating systems with their hardware.

Today, Windows’ familiarity is probably its greatest virtue. Goodness knows, as an operating platform, Windows leaves a lot to be desired, especially in terms of stability, interoperability, and, in the case of 95/98/Me, security. But we’ve learned to work around many of its flaws, so we feel comfortable with Windows in spite of its problems

Microsoft, of course, does all it can to protect its niche. One of its most powerful weapons is the pre-announcement. The company, often years in advance, announces a new product or a new version of an existing product, and tells how it will work. It then delays and delays, and when the final product comes out, many of the features originally planned are postponed until a later version, or simply disappear. Meanwhile, corporate purchasers delay committing to released software from competitors because they want to see what Microsoft will come up with, figuring that applications from the operating system vendor will be better integrated with Windows.

It’s easy to get stuck in the rut of the familiar, but it’s prudent to broaden your horizons.

Listen to the members of your organization who want to pilot new systems and applications. Give them the freedom, the resources, and some time to investigate whether new products would meet your business needs better than existing systems. Remember, someday you’re going to have to upgrade, and if you haven’t done your homework beforehand it’s very easy to continue down the path you’ve always trod, sidestepping the familiar bumps and puddles.

I’m not trying to minimize the importance of migration issues. They’re always going to be an issue when you choose to change applications or platforms. So is price. But there are times when it makes sense to make changes despite those issues.

If you’re a CIO, don’t let your IT department off easy. Reader Brian Larson, in an e-mail message he sent me about my previous column, expressed dismay about the many times “platform standardization decisions become decisions made to benefit the support system, not the user who needs to get the work done.” Don’t let that happen in your organization.

Make sure you’re as familiar with your alternatives as you are with your existing systems and applications. Reader Mordechai Ovits observed, “Microsoft has inertia. Linux has momentum.” I’m looking forward to seeing whether one is an immovable object or the other an irresistible force.

My way or the highway

Certain proponents of particular operating systems are like adherents to some religions — their way is the right way, and don’t you dare attack it. Or, similarly, someone else’s way is wrong and offensive, and they feel compelled to tell you so — and why — at great length.

These operating system disputants tend to be advocates of fringe operating systems. A few years ago, if I had written an article that, in someone’s eyes, slighted OS/2 — regardless of whether that was the main intent of the article — the OS/2 proponents would let me know how wrong I was, in no uncertain terms. Usually their messages impugned my motives — “how much is Microsoft paying you to write that?” — instead of addressing the points of contention in the article.

For years, the Macintosh has been marginalized as a fringe platform within IT organizations. IT departments like to standardize on certain hardware platforms and operating systems, and Macs are not the platform of choice for most shops. Generally, in companies that have standardized on PC platforms, it’s only graphic designers who get to keep Macs. And that’s a measure of the designers’ determination to continue to work on Macs, when virtually all major professional design software is available for PCs.

Mac users tend to be more relaxed about their status. Maybe it’s a holdover from Apple’s counterculture roots. Remember the computer “for the rest of us” and the 1984 Super Bowl commercial? Apple has always tried to appeal to a mellower crowd.

However, there are still a few groups that rise up shouting when their favorite platforms come in for criticism. Two that spring to mind are users of Linux and the Palm OS. (I’ve never written about BeOS — I wonder if its devotees have that missionary fervor.)

Linux and Palm OS are both good operating environments, but both have distinct drawbacks. Most expert users of both platforms, if they took a dispassionate look, could probably outline those drawbacks better than anyone else. But the moment they hear a discouraging word, some of these OS partisans get lathered up and have to fight back — even if no one was really challenging them.

That kind of response is counterproductive to these users’ causes. Rational discussion of pros and cons can lead to incremental improvement in products. But when extremists attempt to bury that discussion under a flood of peripheral issues, they quash the opportunity for improvement. They also weaken their case among dispassionate observers; if their favorite operating system can’t stand up to constructive criticism and reasoned discussion, it must have flaws that such discussions would uncover.

The trick for IT departments is to figure out what tool is right for a particular job, and use that tool. In some organizations, particularly centralized organizations that have standardized on certain hardware and software, that task may be easier said than done. In those cases, sometimes it takes a zealot to prove that the unconventional choice is the right one. That’s fine — but confine the arguments to the specifics of the situation, and don’t bring up arguments that degenerate to “Mom always loved you best.”