Comment by JoshTriplett | Hacker News Reader

Comment by JoshTriplett | Hacker News Reader

Parent Full thread

Even among words that do theoretically have an ordering preference, some people have different habits. I knew someone who habitually said "three or two", rather than "two or three".

Leaving aside the effect of prosody mentioned in another comment, I think the rest of it is habit, together with the brain's tendency to group things. If you're used to hearing "plus or minus", your brain may be grouping it together into a phrase whose meaning you understand directly without decomposition, so if you hear "minus or plus" there's a moment of having to map the components to the composite meaning, together with wondering if the speaker intends the difference to have a meaning.

(In mathematics, there's a reason to have both ± and ∓, since you can use them both in the same equation when you need "minus when the other one is plus, and plus when the other one is minus". For instance, a±b∓c means "a+b-c or a-b+c".)

I don't know if this has a name, but if it does I'd love to know it.

> For instance, a±b∓c means "a+b-c or a-b+c".

I feel like this would be less confusing if rendered as "±(a+b-c)". It's the same amount of cognitive effort for figuring out the second term/factor, but you get the first one for free.

What you’ve suggested is not equivalent. The a is always taken without negation, and that’s not the case in your rendering.

Ah. "a±(b-c)" then.

The fact I made that mistake in the first place just makes me more convinced the original form is unnecessarily complex to parse.

You're too fixated on the example. It's obviously used in cases where you cannot group like that:

    (-b ± sqrt(b² - 4ac)) / 2a = 2c / (-b ∓ sqrt(b² - 4ac))

that makes a and -a

"three or two" means "likely three, maybe two"

"two or three" means "likely two, maybe three"

Here, the ordering is significant and has nothing to do with a habit.

That's not a common connotation; I've never once seen anyone use those phrases to mean that. "two or three" does not typically have any connotation of either being more likely.

Might be something specific to where you are?

What I described is how this construct is used in every place I lived.

Have you ever lived in the UK? Because I've never thought of "two or three" as meaning that two is more likely than three.

Are you sure that other people actually mean this when they say it, or are you just adding the extra connotation in your head?

Also I'm not sure I've ever heard anyone say "three or two". That sounds weird, like saying "fork and knife".

Lest anyone draw the conclusion that this is a US thing, it certainly is not. I've lived in many places throughout the US and have never heard anyone say "three or two". Nor does "two or three" carry the connotation that two is likelier. A closer fit to that would be "a couple", which formally means two, but sometimes means "a small number".

Not in the UK, no. But I did live in Canada, the US and across Europe.

Not that it's commonly used and not with 3 and 2 specifically, but when used and the pair is inverted, it's meant to convey that the former is more likely. That's just like ... common sense, really.

Reply on news.ycombinator.com