Received Wisdom on Risk: Internet Explorer and Firefox, Flying versus Driving
Oh God! Not another article about the evils of Internet Explorer? Actually, no. What interests me is not what brand someone prefers, but the gap between perceptions and realities, the gap that makes people afraid to hop in a commercial aircraft, but not afraid to hop into a car, even though the aircraft is much safer on a per-mile basis.
You’ve probably been told a million times that it is much safer to fly than to drive, but have you heard that the car is actually safer on a per hour basis? I hadn’t until just recently, but it seems as though it may be true based on some numbers. Those number suggest, though, that flying is more dangerous even on a per-mile basis. Could that be?
Well, as with most such things, it depends on the assumptions one make. If you try to make the comparisons as usefull as possible by comparing similar trips, things look a little different. So there are really two questions:
- Is it safer to hop in a plane or to hop in a car on average?
- Is it safer to take a given trip in a car or a plane
Our typical decision is not to decide between taking a trip up to the mailbox (which is two miles away for me) or to hop in a plane. Rather, when faced with a trip of say 1000 miles do I drive or fly? According to academic statisticians Michael Sivak and Michael Flannagan, writing for American Scientist on the relative safety of flying and driving, things tip dramatically in favor of commercial air travel (general aviation is much riskier) when similar trips are compared. In this case, they assume that drivers are going long distance on rural highways in non-commercial cars and trucks which, as it turns out, is the safest driving situation. But the make the reasonable assumption that these will be long trips.
As it turns out, based on their assumptions (which I find reasonable despite comments from critics), driving is only safer over very short distances. Basically 18 kilometers of driving represents the same risk as one takeoff and landing in a commercial airliner, which is the relevant comparison, because 95% of airliner fatalities happen during those times. For a distance of 1157 kilometers (average length of a non-stop flight, again, a reasonable way to figure this), they estimate that it’s about 65 times as risky to drive as to fly. What does that mean on a per hour basis? Since the average cruising speed of an Airbus 320 is 863kph, if we divide that by 65, we get 13.7kph or about 8.5mph. Let’s round that up to 10mph to account for the time it takes the aircraft to get up to cruising speed. So based on those numbers, for driving to be safer on a per hour basis, we would need to assume that average driving speeds are less than 10mph, which might be true for commuters in the Bay Area, but not generally.
So though I keep hearing lately that it’s safer per hour to drive and safer per mile to fly, that may not be true. I also keep hearing that it’s safer to take a short trip in a car than a long trip in a plane, but never imagined that the risk breakeven distance would be 18 km. And it is perhaps worth pointing out that it is much safer to take a short trip than a long trip in general, and safer still to take no trip at all unless you live in a particularly dangerous place. Then it’s safest to take a one-way trip no doubt. It’s probably also worth point out that the environmental cost of a short trip in a car is much lower than the cost of a long or short trip in a jet.
So what does this have to do with Internet Explorer and Firefox? Was I digressing again? Not possible. I reject the whole concept of digression.
You have no doubt heard that Firefox is safer than Internet Explorer because Microsoft produces such crappy products that are so full of security vulnerabilities. The proof for this the constant announcements of this or that virus that is exploiting a hole in Internet Explorer to do malicious things to your computer. Makes sense, except for a not-so recent computer security study that I came across which shows, among other things, that Firefox 1.5 had twice as many publicly-documented security flaws as Internet Explorer 6 SP2. Now, the Firefox people might say they are more open about their flaws or that the patch them quicker, which could be. Anyway, it doesn’t make any difference, because as the researchers found out, Internet Explorer actually is more dangerous to use, but not because it is poorly built, but because it is so popular that it is targetted much more often.
So in this case, the conclusion is unchanged, but the reasons are more complex than Micro$oft $ucks as the saying goes (personally, I cringe when I have to use IE6 or earlier, but IE7 is almost as good as FF2). What I find interesting, and what I’ve been thinking and reading about a lot lately is how this all relates to the decision-making process. What assumptions do we make? Are our decisions based on verifiable data and are the data based on reasonable assumptions? Not typically, and that impacts our day to day lives on issues from Iraq to global warming. If I ever win the lottery, I’ll devote my time to writing about decision making. After all, winning the lottery is just a matter of getting your Lucky Number System down and working it right? I’m more likely to win the lottery than to get struck by lightning, right?
Popularity: 6% [?]
Add to: 
del.icio.us •
furl •
reddit
•
digg •
technorati cosmos
