The trouble with being a former typesetter is that every day online is a new adventure in torture. Take the shape of quotation marks. These humble symbols are a dagger in my eye when a straight, or typewriter-style, pair appears in the midst of what is often otherwise typographic beauty. It’s a small, infuriating difference: “this” versus ^aEURoethis.^aEUR
I’ll stop replacing curly quotes with straight quotes on OSAlert the day the tech industry gives me back my Dutch quotation marks (^aEURzLike so^aEUR, he said) and adds multilingual support to Google Now and Siri and so on (which right now require a full wipe to change languages, making them useless for hundreds of millions of people who live bilingual lives).
Yes, I can be petty.
Tensions are very high in the world.
There is a high risk of yet another war.
Possibly a big one.
But yes … let’s quibble over quotation marks …
Edited 2016-12-28 22:56 UTC
Here’s how Thunderbird’s built-in RSS reader rendered that:
Also, while curly quotes would definitely be nice, it’s common practice to curse CMSes like WordPress for mangling code snippets by introducing them (or em-dashes) because, somewhere along the way, communication broke down about whether or not they were desired.
In my Thunderbird it looked like this:
Ergo it’s the typical “re-encode UTF-8 in ISO-8859-1” I’d expect from American/English websites:
Usually I don’t bother using the proper ^aEURzGerman quotation marks^aEURoe, even tough it’s not that hard on Linux to type: AltGr+V = ^aEURz and AltGer+B = ^aEURoe, but I would expect any paper, news paper or book to use the correct quotes (could also be ^A>>this^A<<, if you ask me, which is AltGr+Y / AltGr+X).
Or ^A<> in french.
We use that in spanish too, no way to propery type them on Windows.
Alt + 174 and Alt + 175.
French-Canadians have the same problems.
For years Google tried to force the AZERTY keyboard layout on us (we use qwerty) when setting up an android device to the French language while in Canada. It only got sorted out on KitKat I think.
The default keyboard layout in Windows 10 is still ^aEURoeCanadian multilingual^aEUR instead of ^aEURoeFrench-Canada^aEUR even tho I haven^aEURTMt seen one of those in years. Yes, even on a laptop with a different physical layout.
I ordered a Thinkpad E570 last month and even tho I ordered on the French-language Canadian site, and selected a French OS with a proper FR-CA keyboard layout, Win10 was still configured to use the wrong layout (the same one Apple use for French-Canadian, which is also different from the multilingual keyboard used by government offices). Can^aEURTMt a piece of hardware like a keyboard tell the OS what layout it have ?
I think laptops all come with dual-layout keyboards in my province (unless custom-ordered like the Lenovo ones) but people increasingly buy cheap english layout keyboards in bulk so they have a hard time typing proper french. This is leading people to be lazy and relying on Word to auto-correct ligatures and other accentuated characters that don’t appear on the keyboard.
I see a lot of people here (in Quebec) using the wrong quotation marks in French because that^aEURTMs what they see on their computers. Lots of people just gave up on ~A§ because they don^aEURTMt know how to type it. Too many keyboard standards, ill configured OS, ISO vs ANSI physical layouts, etc^aEUR| Technology is supposed to help us, not lead us in a cultural race to the bottom. Next thing you know, people won^aEURTMt even remember when it was possible to type a ligature. Or that those are actually a thing.
I’m in Ontario and my solution, even before I started actually caring to learn French, was just to take advantage of Linux’s support for remapping RightCtrl to Compose.
Easy to use and easy to remember the mappings on my used pre-2013 Unicomp Classic 104 with monolingual QWERTY keycaps.
It also lets me use spellings like “na~Ave” which I think make more sense than the official English ones. (It’s not supposed to be pronounced “knive”, Microsoft. Support Compose so we can use a diaresis.)
In Romanian we have similar quotes, like ^aEURzthis^aEUR. I used to not use these, and I used to not even use the diacritic marks (~Ac"Af`ETM`E>~A(R)) that used to be harder to get right on mobile systems.
It’s not the case anymore – two or three years ago I switched to using all diacritic marks properly, and this year I started using the ^aEURz^aEUR and ^A<<^A>> properly – as Romanian requires it. It’s possible because there is a ^aEURzRomanian for Programmers^aEUR layout that really allows me to use a standard US layout with ALT-GR as modifier for all the special marks.
It’s hard, but someone has to push it. Why not the IT people themselves?
Yup, the ^aEURzprogrammers^aEUR version of the romanian layout is quote awesome.
I’ve started to use proper diacritics even when writing email.
What I’ve noticed is that, given the correct locale settings, LibreOffice Writer will put the correct type of quote when pressing the “ key before the quoted text and after.
The article mentions also dropping diacritics in Polish. As a Swede this baffles me. Dropping the ~Ayen, ~A¤ and ~AP would just be plain wrong. I cringe when I see “sm~APrg~Ayensbord” spelt “smorgasbord” even if it’s an English text. We’ve always had it quite easy though. The extra letters were available on most platforms and are few enough to get their own individual keys on the keyboard. By contrast I know that Romanian has had an especially hard time getting proper support for the `ETM and `E> letters.
Out of curiosity, what’s it like having to deal with a language and not be able to use accented letters? How much harder does it get to read? Romanian doesn’t seem to be as heavy on the diacritics as the Slavic languages ^aEUR“ I guess they’d have a harder time.
As a small personal anecdote: I had a Spanish teacher that didn’t know how to produce the Spanish ~A±, ^A! and ^A? on the keyboard and would add them to the printed document using a pen. It was a language class so leaving them out was not an option. It did look awful though.
I really would like to see diacritics more broadly used instead of finished and the reason is very simple, on languages with Latin ascend, like Portuguese, French and Spanish it helped create a consistent bridge between phonemes and syllables and as so if you could read/write a word properly you very likely could spell it, instead of relying on oral tradition, like it is common on English.
I suspect, the reforms that happened on many places were advanced to make room for an easier learning for natives and foreign, I’m not sure if it is really desirable as it accelerates the gap between old texts and current ones that already build up as years come and go. Many students can not properly interpret texts that are only 100 years old so, if this urge to simplification keeps going, what will them be able to read on 300 years? Fewer and fewer texts or “translated” ones with all shortcomings they introduce. It is not a good prospect.
Edited 2016-12-29 12:28 UTC
In Poland it’s pretty common to see people not using diacritics, especially younger people, or the same kind who don’t use punctuation marks. It’s seen as lazy and is looked down upon, but no, it’s not very hard to read. There are few cases where words may be mistaken for one another, but, based on context that rarely happens.
Historically, back before UTF-8 was prevalent there were two encodings used for Polish. Windows had its cp-1250, *nix systems used iso-8859-2. That caused a lot of problems. For example: on IRC (purists that they are) many Polish channels outright banned Windows encoding. Most Windows users, instead of looking for complicated workarounds, simply dropped Polish accents altogether.
This is still not fully resolved, ’cause now we have three encodings instead of two
Want to to download Polish subtitles for that movie or a TV show? Well, which encoding (if any) is supported by your device is pretty much a lottery. Also, when I played World of Warcraft around five years ago it did not allow me to type Polish letters in chat in any way. I think AltGr+a combination ("A…) used to delete the whole line…
That said, modern smartphones pretty much force people to write accented letters by means of autocorrection. I wouldn’t worry about them disappearing.
For Romanian it’s somewhat easier since the words are not easy to miss when you have the proper context – as a matter of fact, most Romanians got used to writing without diacritics and to reading text without diacritics. After a while everyone gets used to it – of course, it makes learning Romanian hell (for non-Romanian speakers) because diacritic marks are important for pronounciation (obviously), but that’s another story.
Funny part is that not even the teaching system doesn’t insist on using proper Romanian – but that’s just a sign of a depressed people and culture. A different topic for a different day, I think.
Regarding the proper support for the right `ETM and `E> it was a thing mostly due to bad standards implementation and piracy (people used old version of stolen software) – but now that’s more or less solved – most of Romania is on updated software, so there’s no reason to complain about the proper `ETM and `E> – only old text and old habits that need to be updated (they used to use the cedilla version, like A"Y, which is turkish). But nowadays most fonts have the `ETM and `E>, not only the cedilla version.
Edited 2016-12-30 10:53 UTC
Interesting… (I there are 2 Romanian-Americans where I work and I’ve met 1 other at a job site). I lived in Brazil for a few years and we used the alt-gr pretty heavily there as well… otherwise you learn the alt codes.
I communicate in Spanish and Catalan so I also need some diacritics not present in a regular US keyboard. And I am also a bit of a geek when it comes to proper punctuation.
My choice has been using the Compose key (https://en.wikipedia.org/wiki/Compose_key) on X11 Systems where you have sections like this:
XCOMM Quotation marks
<Multi_key> <less> <less> : “^A<<” guillemotleft # LEFT-POINTING DOUBLE ANGLE QUOTATION MARK
<Multi_key> <greater> <greater> : “^A>>” guillemotright # RIGHT-POINTING DOUBLE ANGLE QUOTATION MARK
<Multi_key> <less> <apostrophe> : “^aEUR~” U2018 # LEFT SINGLE QUOTATION MARK
<Multi_key> <apostrophe> <less> : “^aEUR~” U2018 # LEFT SINGLE QUOTATION MARK
<Multi_key> <greater> <apostrophe> : “^aEURTM” U2019 # RIGHT SINGLE QUOTATION MARK
<Multi_key> <apostrophe> <greater> : “^aEURTM” U2019 # RIGHT SINGLE QUOTATION MARK
<Multi_key> <comma> <apostrophe> : “^aEURs” U201a # SINGLE LOW-9 QUOTATION MARK
<Multi_key> <apostrophe> <comma> : “^aEURs” U201a # SINGLE LOW-9 QUOTATION MARK
<Multi_key> <less> <quotedbl> : “^aEURoe” U201c # LEFT DOUBLE QUOTATION MARK
<Multi_key> <quotedbl> <less> : “^aEURoe” U201c # LEFT DOUBLE QUOTATION MARK
<Multi_key> <greater> <quotedbl> : “^aEUR” U201d # RIGHT DOUBLE QUOTATION MARK
<Multi_key> <quotedbl> <greater> : “^aEUR” U201d # RIGHT DOUBLE QUOTATION MARK
<Multi_key> <comma> <quotedbl> : “^aEURz” U201e # DOUBLE LOW-9 QUOTATION MARK
<Multi_key> <quotedbl> <comma> : “^aEURz” U201e # DOUBLE LOW-9 QUOTATION MARK
Excerpt from: https://cgit.freedesktop.org/xorg/lib/libX11/plain/nls/en_US.UTF-8/C…
In Android I use AnySoftKeyboard (http://anysoftkeyboard.github.io/) which have a plethora of quotation marks hidden under the quote key.
Edited 2016-12-29 09:08 UTC
Tbh, I write in Greek and English most of the time,
and while it’s easy to just use the common quotation marks (“) in both languages, when I learned how to insert the left and right double angle quotation marks (^A<< and ^A>>) used in printed Greek, I started using those, in either Windows, Linux or Android, it felt more familiar.
Edited 2017-01-01 02:23 UTC
For me they look better. They are also universal in that regard they work fine in programming also.
Agreed, having multiple quote indicators is a prime source of security vunls. Just use single and double quotes as prescribed by ascii. Standardization is a good thing.
You can really annoy your fellow programmers by using visually similarly looking characters like singe and double quotation marks instead of regular ASCII ‘ and “, and you can even use the many different kinds of hyphens and spaces, not to mention the many “lookalikes” for letters. My favorite: Make the C / C++ / C# compiler choke on what you think is a semicolon – but it’s actually a greek question mark…
And word processors have added them in languages where they don’t belong. Back in school many of my class mates would hand in assignments containing English style quotation marks: ^aEURoes~Ayenh~A¤r^aEUR instead of proper Swedish ^aEURs~Ayenh~A¤r^aEUR. I never used Microsoft Word myself, but it seems like at the time the default setting was to automatically insert English style quotation marks regardless of language.
While the smart substitution of quotation marks does seem to work for English, it often makes a mess out of apostrophes. I bought two books recently where the apostrophes can take any of these forms with no discernible pattern: ‘, ^aEUR~ and ^aEURTM (straight, left curly and right curly). It’s a once you see you can’t unsee type of thing. They were self-published print-on-demand books though. To those of you who write a lot in English and care about proper typography: how do you make sure you get the right character for the apostrophes?
Finally, I wish people would turn of smart quotes when they post source code.
B~A¤r v~Ayenrt f~Ayenr ~APver ~Ayenn p~Ayen ~APn.
Edited 2016-12-29 12:30 UTC
Use LaTeX.
(please notice how on OSAlert we use a whole different way to quote! and how that does show only an opening curly brace)
Very nice overview on the same topic:
http://practicaltypography.com/straight-and-curly-quotes.html
Edited 2016-12-29 14:51 UTC
It’s so obvious why people don’t use proper quote symbols – at least with US keyboards. They are impossible to type!
Seriously, how do I type them?
It’s even worse than I thought! (especially on Windows) https://chrisbracco.com/curly-quotes/
CaptainN-,
Was he being sarcastic? Perhaps what he should have said was ^aEURoeIt turns out that producing curly quotes on a computer is ^aEUR~super easy^aEURTM. I^aEURTMve committed these ^aEUR~simple^aEURTM shortcuts to memory.^aEUR
BTW, I actually find it easier to copy/paste those quotation marks from another source (like charmap) than to type in the unicode ordinal number. Especially if you have a lot of unicode characters to memorize this way.
HTML Code: https://dbaron.org/www/quotes
^aEURoeThis sentence is surrounded by &ldquo and &rdquo , which are a type of quotation marks.^aEUR
^aEURzThis sentence is surrounded by &bdquo and &ldquo, which are a type of quotation marks.^aEURoe
Suprised about the simplicity of markup languages to resolve the problem. But this layer SHOULD BE invisible to the multi-linguist typewriter.
A function key… When presed togheter with quotation key, a contextual map appears around my typing cursor. So pressing the right key surrounding the quotation key, bring the CHILD quotation I was thinking of.
Pressing [Shift] twice would enable Shift-Lock -at my FirefoxOS kb. No need of additional keys. We could use [Alt] twice, as ex.
Edited 2016-12-29 22:02 UTC
All printers printed Latin. Not the technology to blame. But lack of market incentive. Users should insist on Universal fonts and easier multi-language switching.
As Thom Say, see a lot more of mixed language documents.
In the beginning, printers – at least, movable type printers – printed in German (At least in Europe)
Which is why people think Old English speakers used to say “ye” – English used the letter ~Az (called thorn, pronounced th), but since movable type printers were typically bought from German makers, they only had German letters, so those printing in English often substituted the letter ‘y’ in place of ~Az.
Eventually, the ‘th’ became the substitution of choice, and that’s what we use for the ‘th’ sound to this day.
At least, that’s the case with English.
I faintly remember writing Dutch quotes on primary school, thirty years ago. But I think they disappeared before the rise of internet. Probably the typewriter is to blame. Nowadays, the computer with graphical displays allows us to write virtual any accent you would need. In that sense, internet is helping in the right direction, since it pushes for aesthetic pleasing displaying of text.
Now it is just a matter of users taking the effort to do so (which I do at least for names of people).
The beautiful example of Drumhellar below shows that the phenomenon itself is by no means related to internet. The appearance of glyphs has always been heavily influenced by the medium on which they were written and by the tools used for writing.
This is true for cuneiform (https://en.wikipedia.org/wiki/Cuneiform_script) that was printed with a stylus in clay, the Nordic runes (https://en.wikipedia.org/wiki/Runes) that were cut in wood and the various medieval scripts that were to be written using a goose quill.
And this goes on. For the Dutch and German readers: the touch screen makes the English disease epidemic. (In this context the English disease is the habit of writing composed words in Dutch or German separately as it is done in English)
Edited 2016-12-29 22:14 UTC
I think the real killer is good old Ascii and what our keyboards map 1 to 1. When developing and copying and pasting code those directional quotes really get in the way. Also as Ascii is far more standardized than many of the fonts so things change around.
If keyboard linked to the different Ascii quote direction then we would probably use them more. It would probably save a lot of problems in programming strings as nesting quotes will be far more parsable.
Even worse than the demise of curly quotes is when quotes – of either kind – are used incorrectly. I’ve lost count of the number of occasions when books, newspapers or online pieces use a quotation mark to denote the start of a quote only to then forget to close the damn thing with another quotation mark.
Ha, this reminds me that I have just started reading the Thomas Covenant Chronicles, and it uses the older style where it’ll have paragraph breaks, but won’t do end quotes where the same person is speaking. “For example, I’m talking as a different person here.
But what I am say as a different person on different subjects still shouldn’t waste an extra quote, if I’m still typing as that person.”
Also, I’m not sure if it’s the reader I’m using or not, but it’s formatted so that paragraph breaks are double spaced instead of indented. Guess it may just be a matter of my settings.
Anyhow back to reading the book, at least it’s not using curly quotes, it just happens to have a weird font I set.
Haha, I had forgotten about the proper Dutch quoting style. Are you sure that’s still being taught in schools these days? I wouldn’t be surprised if even the teachers now use “”.
According my daughters proper Dutch quoting is not taught anymore. O tempora, o mores.
In American English we ALWAYS have the periods and commas inside quotes while Question and Exclamation Marks are inside or outside based on context. This illogical rule governing periods and commas was based on a limitation in typesetting technology of a couple hundred years ago. The Brits fought back and won.
You have no idea how much of your day is governed by limitations in your technology or bureaucracy. Why do I have to name the baby before leaving the hospital rather than getting to know them first like most cultures have done for eternity? Why do web forms have first/ last name rather than surname and given name? Why doesn’t the time keeping software bump your day off to the Tuesday after the Monday holiday rather than the Friday before which could put it into a different pay period? Some people could fall into a Leave Without Pay status based on poorly written software.
In this case, the displayed language is less readable (where does the quotation begin or end?) due to an outdated limitation in old typewriters. Great example for Systems Analysis or Project Management of how bandaids and workarounds in the old systems become established in the mindset and are unnecessarily carried over to the new.
On the other languages front:
Software defined keyboards with LEDs (or some display tech in each key) would fix many of these issues. Obviously there are many non-American English or bilingual KB layouts that can be met and a more versatile/ flexible KB layout would be great.
Just a quick search:
https://www.amazon.com/Optimus-Maximus-OLED-keyboard-white/dp/B0042F…
There is demand for something that does this, but at a cheaper price point.
Probably because little effort to teach people what those terms mean, so people who aren’t familiar with foreign naming conventions choose “first” and “last” because it’s “more intuitive”.
I’m still amazed how long it took me to learn those when I was a kid… though it doesn’t help that you’re using “surname” rather than “family name” and some people use “christian name” in place of “given name”.
(I still see “surname” infrequently enough that, whenever I do see it, I have to mentally cross-reference to refresh my memory of which one it is by pulling up instances when someone said “surname and …” or “… and surname”.)
Bureocratic Expediture has made a handful of surnames to dominate directories, at my Country. Rendering those helpless at ID Us. That’s also technology. Thanks, Hae-You. [Sorry about lacking English Correcting tools at this little device.]
Can’t tell you how many times non-standard quotes have caused me hours of my life because while they would show up as normal quotes when you copy and paste them from somewhere into a text file, they are still interpreted as the non standard quote, breaking things.
They serve no real purpose but to piss off coders. If you want them, use a different damned font.
So you’re saying the whole internet should cater to coders only?
The whole Internet’s base consists of coders’ code, and it runs on (and because of) coder’s code. The whole user content is just a filling inside coders’ code.