Conventional IT wisdom

[This week's Frankly Speaking is a rerun because I was out sick with a busted eardrum. It originally appeared in Computerworld on Sept. 25, 2000.]

A few things in IT remain true, despite the best efforts of vendor marketers, management gurus, industry analysts and Computerworld pundits to make everyone believe otherwise:

The Internet poses security risks. Always has. Always will.

Users don’t understand IT — never have, never will. All they really understand is their jobs.

No project gets enough time, budget and resources to be done the way it should be done.

"Free" anything isn’t.

Faster hardware is cheaper than faster software.

Vendors and consultants are trying to make as much money from customers as possible. It's up to us to get our money's worth. Caveat emptor.

The best technology doesn't always make a successful product. Then again, the best technology may not be what you need.

Some vendors really don’t like some other vendors — so much that they're willing to let it get in the way of working with customers.

If nobody else is trying something, there's usually a reason. Maybe not a good reason, but a reason.

Faster hardware doesn't solve business problems — unless the business problem is slow hardware.

The CEO will always think consultants' ideas are good because he's paying good money for them.

Traffic expands to fill the bandwidth provided.

If you take something away from users, they'll sneak it in the back way anyhow.

The most powerful influence on CEOs' IT preferences are the people who write for airline in-flight magazines.

"More bandwidth/memory/storage/processing power than you'll ever need" will last you six months. A year, tops.

"We've never done it that way before" is a more powerful argument than any cost-benefit analysis.

IT projects advance or die. Sometimes both. But if a project isn't advancing, it's dying.

Nobody ever got fired for buying the flavor of the month.

What counts isn't how much a product costs when you buy it. What counts is how much it costs before you finally shut it down.

Functionality isn't the same as usefulness.

When you just have a hammer, everything looks like a nail. Most IT people just have technology.

It always takes longer and costs more to fix it later.

The systems that last are the ones you were counting on to be obsolete.

A good idea is no match for a bad habit.

By the time your CEO has read about a technology, it's no longer a strategic advantage.

Ninety percent of a system's cost is still training people to use it.

IT projects fail. Large projects fail more often than small ones. So if failure isn't an option, you'll never do anything.

If you think your company's users are awful, just wait till you’re on the Web and have customers of your own.

Exactly what you want always costs more than what you can afford — whether it's technology or IT employees.

Old ideas got that way because they proved useful.

Data isn't information. Information isn't knowledge. Knowledge isn't manageable.

Systems aren't made from metaphors, paradigms and methodologies. They're made from code, wires and hardware.

The Model T didn't become a standard because it was the best. It became a standard because it was the cheapest.

The hardest problems get solved last.