Comment by Voklen | Hacker News Reader

Comment by Voklen | Hacker News Reader

I do often wonder whether we should maintain traditional typography when moving to a digital age because punctuation evolves as language does. If we’ve deemed it unnecessary to have seperate symbols for each of the dashes and everyone uses language that way then that’s fine. We can also ask this question about smart quotes, you’ll notice I’ve been using the U+2019 as the apostrophe here and I could “quote” like this. It's a question of how much ambiguity it causes, how easy it is to input, and how subjectively aesthetically pleasing it is.

My personal opinion for hyphens is:

- Ambiguity: most can be cleared up with spaces, and for examples like 3-8 if it’s numbers we can tell it’s a range from context

- Ease of input: one character is a lot easier to decide between than 3 (or 4 if you include minus), and if there are rules for software to be able to input the correct character every time then the differences in characters become redundant

- Subjective aesthetics: I quite like the consistent compactness of the single hyphen

And for quotes:

- Ambiguity: They show when quotes start and end which is quite nice and we can have nested quotes. But these are things that are not critical to meaning and simply make it easier

- Ease of input: Usually automated but can absolutely tear through code if pasted in the wrong place. If we deem these smart quotes useful enough then they can coexist with typewriter quotes peacefully if we do not run the quote formatting on code blocks (which is where code should be anyway)

- Subjective aesthetics: I do like the look of smart quotes but would be willing to use straight quotes

The pragmatic thing is to stay glued to the typewriter and then escape our nested strings with Unix toothpicks everywhere.

> Ambiguity: They show when quotes start and end which is quite nice and we can have nested quotes. But these are things that are not critical to meaning and simply make it easier

Typographic conventions go further than that.

In Norwegian it’s `«»` for one level of nesting. For nested quotes you are supposed to use something else. Maybe `‘’` (single quotes) for the second level and then `“”` (American English double quotes).

Maybe American English uses `“”` and then `‘’`.

In my opinion that’s not necessary. At least for text storage.

Part of my complaint about that is that although I think the different punctuation marks are great, using them is a pain because of keyboard layouts.

It's easy to find a hyphen (or something close enough) on your physical keyboard, but there's no em dash. OSes also make it a pain to automate even when they claim otherwise.

I go out of my way to use em dashes but do I think others would? No way. So is lack of use because of lack of utility or because of idiosyncrasies in keyboards?

Hyphens are great for some things but are too short to visually offset text.

The Mac layouts handle the dashes well in my opinion (quotes not so much). Option+‘-’ is ‘–’ (en dash), Option+Shift+‘-’ is ‘—’ (em dash). Option is equivalent to AltGr in the Windows PC world.

What about using tilde for numeric ranges?

"The global conflict spanning the years 1939~1945 is known as World War 2..."

Ohh yes, reducing the amount of tasks the hyphen is used for helps as well

Tilde is already used for approximation though.

The sentence as you wrote it could be misinterpreted as "the conflict spanning the years 1939 to ca. 1945...".

Had you used a dash/hyphen/minus/whatever nobody would be likely to misinterpret that as "the conflict spanning the years minus six..."

No, ≈ is used for approximation, ~ is just the most similar ASCII character, and it became ingrained by people used to using old computers. Just like * is not a multiplication sign, but × is.

In other words, tilde is used for approximation just like the asterisk is used for multiplication and “literally” is used figuratively. We can argue over those uses being correct or incorrect, but they are used like that.

Thus I agree that using tilde for numeric ranges would be confusing. Might as well just use a hyphen, which is easier to type and most people won’t notice the difference from the correct character (en-dash).

> but they are used like that.

Using that form of reasoning, it could be claimed that, say, “espresso” is pronounced ”expresso”, because some people do pronounce it like that.

But that would be disingenuous, since “is pronounced” does not generally mean “is sometimes, by some people, pronounced”, but “is supposed to be pronounced” or “is properly pronounced”. The same goes for “tilde is used for approximation”; no it isn’t. If would be different if scbrg had written “tilde is sometimes used for approximation”; it would have indicated a possible interpretation of the first meaning, and not the second.

> If would be different if scbrg had written “tilde is sometimes used for approximation”;

Oh, dear lord. I apologize for leaving out this very important word. I thought it was fairly clear that I didn't mean it was the only symbol used for approximation, pretty much like how, I don't know... nothing is the only thing used for anything.

Whatever phrase, symbol, word or tool in general you find, you can be fairly certain that there's something else that could be used instead.

In the really real world, people tend to use the symbols that are easy to type with their keyboards. Ironically, this is a bit like what TFA complains about; people always use the hyphen that's available with one keystroke when in fact they "should" (for some arbitrary value of "should") use a handful of different ones. And they use tilde for approximation, because nobody knows how to type a fucking ≈. You'll also note that they use " when they "should" have used “, ” or any of the umpteen other variants of quotation marks.

When it comes to ambiguity, which was what this sub thread was about, how things are often used is actually quite important. Because, you know, it's what people actually write that you have to disambiguate, not what they should have written.

OK, fair enough. I was recently on the other side of that same argument here on HN: https://news.ycombinator.com/item?id=34940715

No, they were right—languages change and a single tilde (~) definitely means approximately: https://en.m.wikipedia.org/wiki/Tilde

Most people associate a double tilde with “approximately equal”.

The asterisk is an approximation of dot, not a replacement for x. They just mean the same thing on scalars.

Using × or ∙ for multiplication is, IIUC, a cultural differentiator – just like English uses . as a decimal separator, but many Europeans use , for the same purpose. But in Unicode, × is “MULTIPLICATION SIGN” and ∙ is “BULLET OPERATOR”, and * is more visually similar to × than ∙, so I assume that’s where it originates.

At this point I actually handwrite an asterisk to denote multiplication. If I think about it I know it's "wrong", but I do it anyways.

¿But which tilde? I'm a fan of typographical abuse of the ⁓ swung dash myself.

Reply on news.ycombinator.com