That is what happened, famously, in the Challenger explosion disaster -- in which, as design expert Edward Tufte famously pointed out, key data was presented to decision makers in a format that failed to highlight the magnitude of risk a freezing-weather space-shuttle launch would entail.

And that is what happened this week in Palm Beach County, Fla., where an apparently well-intentioned effort on the part of local officials to make their ballot more readable for elderly citizens led to voting-booth confusion that could well have tipped this squeaker election toward George W. Bush. (Florida, it seems, is ground zero for information-design catastrophes.)

In other words, as software analyst Joel Spolsky asked on his weblog page: "Is Bush going to actually be the next president because of a usability bug?"

The flow of reader comment to us here at Salon is divided between those who are outraged at the perceived problem with the Palm Beach ballot design -- and those who say, as one letter writer did, "Anybody that says it's confusing is basically an idiot. The arrow points directly to the correct hole ... Pay attention next time."

But, of course, it's the nature of design flaws that they introduce confusion around the margins of information, precisely where people aren't paying close attention. It's easy to look at a ballot, once you've been prepped to be on the alert for potential problems, and say, after the fact, "This doesn't look too confusing." It's when you're not on guard -- when you're just repeating a process that you've done many times before (for instance, elderly Palm Beach voters casting ballots) -- that a design flaw can lead you astray.

That's certainly the consensus of usability and design experts on the Web, who have been posting their analyses of the Palm Beach ballot design since it surfaced as a critical factor in the presidential election.

"The Florida ballot clearly had usability problems, caused by the attempt to map a two-column set of labels onto a one-column action area," Jakob Nielsen, the author of "Designing Web Usability," writes on his home page. "A direct mapping between two single-column areas would have been much less error-prone."

Dan Bricklin, the creator of Visicalc and Trellix, agrees. His site offers a detailed analysis of the ballot's flaws, and asks, "I wonder what the usability testing for this was like, or even if there was any such testing."

As for those who cry, "Any idiot should be able to figure out that ballot," Bricklin argues: "Errors on tasks arising from 'dumb mistakes' are very common, with rates of easily 5 percent, 10 percent or more. Elections, even important ones like for president of the United States, are often decided by much slimmer margins than that."

Most of the time "usability experts" get less attention paid to them than they (rightly) feel they deserve. So when something like the Palm Beach fiasco gives them the mike, they tend to grab the opportunity to rail about how screwed up everything is.

Nielsen, for instance, isn't prepared to say that the Palm Beach design tipped the election because, "There were so many other usability problems in other states: All we can really say is that it is obvious that the people who create voting designs don't bother much with usability."

Maybe, after the Palm Beach ballot interface disaster, they'll bother a little more.

The Free Software Project
Read Andrew Leonard's book-in-progress on Linux and open source -- and post your comments.