Hacker News new | past | comments | ask | show | jobs | submit login

A) I was just assuming that your numbers were correct. Regardless, though, the numbers aren't being reported by a scientist. They are being reported to consumers in an advertisement. I wouldn't call leaving off sig figs, or even rounding, sleazy in this case. But we probably just disagree on that.

B) $12.3 isn't a very good comparison because in the context of currency, the numbers after the decimal really signify number of cents, not a fraction of a dollar (although they are equivalent).




I was just assuming that your numbers were correct.

Just to be clear, these are not my numbers, they are Microsoft's numbers. If the weight is really 1.3476 lbs and Microsoft reports it as 1.3, the number is "correct" and they've told no lie, but it is certainly unfair to print the competing number to an extra digit (1.44). On the other hand, if the weight is really 1.3476 and they reported it as 1.30 that would be a flat-out lie (as in "class action lawsuit for false advertising"). So, did they print it as "1.3" instead of "1.30" because zeros aren't first-class numbers, or because the zero would be a lie? As I said earlier, I'm willing to believe that it is probably really 1.30, but none of us know that for sure -- Microsoft has made no statement about what comes after the 3.

Regarding (B), we have "cents" because we consider two digits after the decimal point to be significant in most contexts (not on your U.S. federal tax forms, if my memory is correct, since you are instructed to supply only whole dollar amounts, and not for stock prices that may have many more significant digits than 2 after the decimal). If you put an item on a digital scale it displays either 1.3 or 1.30 or 1.300 depending on how many digits are considered significant. To do a side-by-side comparison where you are treating the 2nd digit after the decimal as significant in one case but not in another is odd.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: