Hacker News Reader

Imagine if, say, a tenth of a percent of cars failed to start in the morning. That would be considered a big deal, wouldn't it?

Enough to employ people to look into the problem(s) and fix them?

View on HN

No, that would be considered shockingly low. There are about 300 million cars in the US, and it's hard to find good statistics on cars breaking down, but AAA makes about 30 million service calls a year. So AAA alone gets the number to 10%, two orders of magnitude higher. (But I don't mean to exaggerate the difference - Facebook accounts must be significantly easier to keep running than cars.)

I think your math is wrong. 30 million a year means about 82,000 a day ("in the morning"). So that's about 2.7 hundredths of a percent. Another way to independently estimate, is that I've been driving for about 7,300 days and had an alternator failure twice that prevented my car from starting. That's...just about 2.7 hundredths of a percent as well. So I think that 0.1% is 3-4 times higher than normal.

Reply on news.ycombinator.com