I love/hate language (Off-Topic)

by iconicbanana, C2-H5-OH + NAD, Portland, OR, Tuesday, May 12, 2015, 03:07 (3279 days ago) @ naturl selexion
edited by iconicbanana, Tuesday, May 12, 2015, 03:23

we usually gave an error of one order of magnitude as acceptable for a correct answer with extremely large numbers,

The article stated "about 1.1%" or about 1/100. That is very close to two orders of magnitude and very far away from three orders of magnitude. I believe you when you say that an error of one magnitude is acceptable for extremely large numbers. This was not an extremely large number though.

The point is that "several" is fairly well understood to mean some value more than two. If you had said "couple", even though some people may assign a value other than two to it, I would have no argument. It fits just fine with the actual information. "Several" does not fit, hence my comment.

I don't really have a problem with your definition of several; I'm gathering the majority of people here don't have the definition I always had.

On the subject of the 1/100 though, I would take issue, mostly because we're not dealing with 1 unit of measure to 100 units of that measure; it's more likely to either be 1.0^30 units of measure to 1.0^32, or perhaps ^-30 to ^-32. The tools they used for the measurement probably aren't necessarily going to be 100% precise here; you could quite possibly see orders of magnitude in variance. You really seem to be clinging to that 1/100 and I don't think it's necessarily productive to be; I was specifically trying not to be precise, which is why I didn't say two orders of magnitude (or a couple orders of magnitude, since I understand that to just mean 2).