back
78 comments
This has been driving me nuts for decades: "Smart Quotes" first appeared as the name of a feature of interactive document editing tools, wherein the user would always keyboard the "vertical" single or double quote ASCII character, and the software would look at the context of where that quote character was being inserted, and choose the correct "left-quote" or "right-quote" character to actually insert into the document.

Unfortunately, users (and product managers of competitors) misinterpreted the label in the check-box in the UI that enabled this feature, and thought that "Smart Quotes" meant that the name of the quote characters somehow had the word "smart" in it. But, no, it just meant "turn on the intelligent logic for which quote to insert into the document."

Source: Me, 1988 or so, working on FrameMaker. I don't think there's a reference to the word pair "Smart Quotes" anywhere earlier than the FrameMaker ~2 manual, and certainly not in any book on typography. I'd be happy to learn that I am wrong, if anybody has a pointer.

I think “smart quotes” is fine as a name, since most people who type them still do it automatically via a context-based logic (whether it’s a word processor or text box input on an iPhone). And it works better than “curly quotes” since they are not curly in some fonts.

If you’re going to be pedantic, I’m not sure that “left-quote” or “right-quote” is correct either. Shouldn’t it be “left double quotation mark” or “right double quotation mark”?

I know you said “most people”, but I always type these typographically correct quotes manually (no automatic substitution or “smart” quotes involved). It's Opt+[ (etc) on macOS, for example. So the name is disorienting for me at least, and as the top comment said, they predate computers and weren't called “smart” then.

But see also https://smartquotesforsmartpeople.com/ :-)

As a Brit I always used to interpret "Smart Quotes" with the old-fashioned definition of smart that some people still use here, i.e. presentable, nice-looking, clean-cut.
TeX (as a mostly batch formatter) didn't do this, but it did map ASCII quote characters to the typesetting ones. Including mapping a sequence of two single quote or backquote characters to the typeset double quotes.

(FrameMaker was great, BTW. It was what got me into developing fancy layout style sheets and templates, which I later also did with Interleaf, MS Word, LaTeX, and various Web stuff. Somewhere in there, I worked on DynaText and DynaWeb at EBT, so overlapped a bit with Frame as it got into SGML.)

Interestingly, the logic of mapping of two single quotes into a double quote was not built into TeX; rather, the metadata in each of the standard fonts simply used the ligature mechanism to handle it, just as they would specify that "f" followed by "t" should turn into the single-character "ft" ligature.

It's a shame that ASCII (and its visual representation on terminals from most manufacturers) has a wonderfully slanty left-quote character, but the right-quote character is typically shown as vertical. This makes TeX input files that contain quotations look kind of yucky, which is all the more galling as the original SAIL terminals had symmetric quotes, so everything looked beautiful there (never mind also having proper uparrow and downarrow characters to indicate superscript and subscript, rather than the ugly caret and underscore characters that we're all stuck with in ASCII-land TeX).

Source: Me, 1980 or so, working on TeX and Metafont.

Words change; terminology changes; language changes. It is the way of things.

In that sense I think "straight quotes" are fine too, although “smart quotes” arguably look a bit nicer, but that's a matter of personal taste IMO.

Related comment from a few months back: https://news.ycombinator.com/item?id=31792626

Thank you. There’s nothing “smart” about my quotation marks; I type them with physical keys just like any other punctuation.

  Left: “
  Right: Shift+“
  Typewriter: AltGr+“
Somehow, my chinese language option still uses smart quotes. For instance,

“一個句子”

This is the first time I'm hearing the term 'smart quotes' or 'soft quotes'. I googled to see what it is and all I got was some 'smart' quotes. There's an obscure Wiktionary page on smart quotes which also does not really explain what it is. The GitHub page being linked itself is so vague and cryptic it's hard to understand what is even talked about.

To all those who are confused, smart quotes or soft quotes are those context aware quote marks that some word processors insert which depending on context, will put either closing or opening quote marks.

They aren't smart or soft, they are typographically correct.

On PC I have AutoHotKey bindings for German and English quotes, single and double, and I'm usually using them.

You might want to give Ilya Birman's Typography Layout a try: https://ilyabirman.net/typography-layout/
On Linux I use Fcitx and Ctrl+Shift+Alt+U to open the Unicode character picker, while on Windows I have my fork of a character picker discovered a while ago (https://github.com/flobosg/character-picker) bound to the same shortcut.
read up on the compose key it may change your life for the better. or it may be difficult to replicate on windows/wayland leading to additional stress in your life. or perhaps you enjoy your current setup, which does not sound half bad, although not as good as the compose key.

Most linux desktops are different and more complicated but here is how I set mine up on openbsd.

pick a key that is handy but you don't use much, for me this was the context menu key, on many european keyboards there is an altgr key set aside for this exact use case.

in ~/.xsession add a line like "xmodmap -e 'keysym Menu = Multi_key'"

create a file ~/.XCompose to setup your own compose sequences. mine looks like

    #load system compose files
    include "%L"
    <Multi_key> <w> <e> <b> : "\xf0\x9f\x95\xb8" # spiderweb
read /usr/X11R6/share/X11/locale/en_US.UTF-8/Compose to get a feel for the existing compose sequences.
> they are typographically correct.

According to an ancient convention.

Conventions can be changed. It is not intrinsically, objectively superior to have two separate glyphs for a starting and ending quotation mark. I'd argue the opposite. It is needless complexity. Let us jettison it and embrace the simpler single glyph.

Convention doesn't mean bad. But sure, feel free to use ( for opening and closing parentheses.
They might be typographically correct, but they are incorrect for code, and cause so many issues. I'd rather have typographically incorrect quotes that work everywhere.
Why would you type them for code? I'm not arguing for the mobile phone automatisms. I insist on separate keys for different quotes.
> There are 126,000+ quotations on HackerNews using smart quotes. I find this deeply disturbing.

Mobile devices appear to do this by default. iOS in my case anyway. I turned them off because it was messing up my hugo blog's frontmatter whenever I start a post on mobile.

You have prompted me to check how to turn it off on iOS. Under settings>general>keyboard, un-toggle "smart punctuation". It also turns off tuning "--" into an em dash. Not sure what else it disables.
Thank you for that, in hindsight I should have chucked that info in!
I just type them in? On a Mac it’s as difficult to type a typographically correct quote as it is to type a dumb quote, just a modifier and a key. OK, so it’s not labelled, but if you know how, why wouldn’t you?
Realised I should have provided some reference (and too late to edit my initial post). On an en-Intl Mac keyboard:

“ = option + [

” = option + shift + [

‘ = option + ]

’ = option + shift + ]

There’s a few others I use regularly as well:

… = option + ;

– (en dash) = option + -

— (em dash) = option + shift + -

Then there’s a bunch of accent meta keys, so you can use option + c for façade, option + e, e for fiancé, and option + u, i for naïve.

If you wonder where the shell commands "library" and "lb" come from, see https://github.com/chapmanjacobd/library

And specifically, "nouns" here are defined by this regex: https://github.com/chapmanjacobd/library/blob/main/scripts/m...

What is a soft quote?
They are better known as "smart quotes"

iOS at the least by default will substitute "dumb" quotes with smart quotes in text boxes.

Why is OP describing that as “disturbing” (I type with “smart” quotes)? It’s not intentional; it’s just because I’m on an iPhone.[a] On the default keyboard, if I want "dumb" quotes, I have to tap and hold the key to get the popup. Simply tapping the quote key is less effort.

[a]: On my computers, I’ll use "dumb" quotes. Easy way to see where I typed a comment from, I guess.

Are there any other places on on the internet that call them "soft quotes"?
Looking around, I got the impression they're double rather than single quotes.
Speculation, but two causes could be: - Writing in a plaintext editor before copying into HN. - Using a third party HN client that only supports plaintext. For example, I imagine there might be at least a few people out there who might access HN from a terminal client?
> Writing in a plaintext editor before copying into HN

Yes, I think it is a thing where people type into Microsoft Word (or similar). Given how moving around the web can cause text in fields to be lost, I don't blame them. But it was a surprising discovery for sure.

Mobile devices is my guess. For example, iOS by default does not use straight ASCII quote, all quotes are "typographically-correct" curly quotes, pretty much in conformance of Apple design-obsessive reputation.
It’s the default substitution done on iOS devices - the OS used on the most widely used smartphone in the US - this is a lot more common among HN users than copying and pasting out of MS Word??
A people using iOS with its default “Smart Punctuation” behavior?
this title makes me roll my eyes. such a blatant attempt at confusing-titleism
Confusing-Titleism, or: how HN bloggers try to trick you into reading their posts
This is kind of sloppy; some of the conclusions don’t follow from the data, because the regexes aren’t carefully constructed. I didn’t read much of it, but in the first one he’s captured all sentences beginning with Intelligent, Intellectual, Intelligence, etc. This could have been made more precise by just including a [. ] after Intel.
> in the first one he’s captured all sentences beginning with Intelligent, Intellectual, Intelligence

This is simply not true. Look at the files in the repo directly

Intel is ^Intel$. The nouns are separated by word boundaries

I’m going by the article, in which they aren’t.
"Sun Ra" might not be the sun god you are thinking of .....

https://www.youtube.com/watch?v=H1ToFXHW5pg

Yes, I know who Sun Ra is. Because I didn't see ^Ra$ in my list of common nouns then it means that there hasn't been a comment just about Ra. But now there has: Ra. Ra. Ra. 4 times total now :-)
Does the “Intel” filter also catch titles with: Intelligen[t|ce]?
Given that

    'sun|moon' 
resulted in

    30905 moon
    36100 Samsung
    44937 sun
my guess is no.
Intel is ^Intel$. The nouns are separated by word boundaries. Feel free to explore the files in the repo