Carrying On: Words to (Net)Work By

• Information is plentiful; time is not. What with Google, blogs and Wikis, there is so much technology information at your fingertips that you can pass as an expert in nearly any subject with only a few hours' research. The hard part is knowing what to pay attention to and what to ignore; this is the value of analysis.

• Throw-away infrastructure is here. Users don't bother repairing their broken DVD players any more; they simply buy new ones. Nor do they analyze Windows when it crashes; if worse comes to worst, they reload the OS from scratch or replace the PC.

Same with networking: When a problem gets to Level 2 support (after Level 1 verifies that the power is actually on), the troubleshooter often just replaces code. The logical next step is to swap hardware. Why pay for expensive equipment and maintenance so a tech can replace a router or switch tomorrow? Buy two cheap devices and use one as a spare.

• Buy only for the near term. Technology is a terrible long-term investment--worse than cars, jewelry and even California real estate. Take it from the companies that bought into ATM networks: Never purchase systems based on a vague future need. When you need the capability, the price will be half what it is today.

• Technologies stay around long past their stay-fresh dates. Few systems are ever really retired. By the time we finally get matter transporters, the people booking our trips will probably be using 3270 terminals.• Businesses cannot restrain consumer technologies. No matter how many corporate policies you write, you will have to deal with home PCs, remote Internet kiosks, camera phones, PDAs and who knows what devices the consumer market thinks up next. You might as well try to hold back the tides.

• When something breaks, it is IT's fault. No matter how much it outsources. IT can't outsource responsibility.

• The best is the enemy of the good. (That's Voltaire, by the way. Obligatory literary reference--Columnist Code again.) By the time you have all requirements, test all products, wait for standards, develop the perfect design and find the best pricing, it's likely too late. The best solution is not the most reliable, or the cheapest, or only available from the largest vendor in the market. Sometimes, the user just needs something cheap, simple, quick--and flawed.

• Respect other people's time. You can give me e-mail, voicemail, instant messaging, a softphone, a laptop with Wi-Fi, 3G, a PDA and a smartphone. Even (shudder) a pager. It might be easy for you to reach me--that still doesn't mean I have time to help you. Just in case, I'll take your IM address.

• Mind the gap. The more your skills are just like everyone else's, the less valuable you (and your work) are. You're just a replaceable cog in the machine. But you'll find job security in the gulf between technology and business.• Surround yourself with people smarter than you. For some of us, that's pretty easy.

• Learn who matters. In this age of matrixed organizations, there are only two individuals you must satisfy: the person who signs your paycheck and the person who can get you fired. Probably, they're one and the same.

• Obey workplace karma. Do favors for people, even if they can't fire you. Eventually, it comes back to you--and often just when you need it. This has saved me my job more than once.

• Remember the people who help you. By my count, there have been at least five major regimes at Network Computing. Their names might not mean anything to you, but the people here mean an enormous amount to me. I have been enriched by their friendship, and I owe them a tremendous debt.

• Know when to stop talking.David Willis is a vice president of Meta Group's Technology Research Services. Write to him at [email protected].