Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

A footnote in the article addresses this: "More logical still would be to have two periods, one marking the end of the quoted sentence and the other the end of the top-level sentence. But that would be redundant and also look ugly."

I tend to agree with the author. While it's logically consistent, it's typographically redundant. There's no need to end a sentence with a full stop, a quotation mark, another full stop, and then a space, before beginning the next sentence with a capital letter.



I don't see why it is redundant. There is no reason to assume that the quoted sentence ends the quoting sentence. Imagine this being the last thing you see on a page. Without turning the page, there would be no way of knowing whether the quoting sentence is complete.

But these things are just conventions. It's futile to demand any kind of logic or consistency.


It's not the only element of style that may seem illogical, there are many more. Off the top of my head:

* Nested parentheses. (In theory, you could use any level needed (provided that they make sense), but in practice, they're scoffed at.) It is similar with quotes, for the same reasons: limited readability - although here you can at least juggle with single, double and French quotes, depending on the language and style guide used.

* Repetitions: unless used as a stylistic device (in poetry, advertising, etc.), your editors will try to modify repeated words, trying to find synonyms. In technical writing, it is an abomination, and competent editors know very well they must not touch any specific terms, no matter how often repeated, as they have very precise meaning.

But in general, the rules present in style guides are meant to ensure consistency and uniform reading experience, so that the reader is not distracted by form and can concentrate on the meaning, so they are a good thing as long as people are aware these are just arbitrary rules separate from spelling and grammar.


In general writing, you have to be careful to balance repetition. At one extreme is what's been called "elegant variation"[0] and at the other is monotonous repetition, which hurts readability.

But, as you say, in technical writing, it's enormously irritating. I do a fair bit of freelance writing, and it pisses me off no end when editors and clients change a technical term for an apparent near synonym because they don't understand the nuance. It makes the sentence nonsensical and makes me look as though I don't know what I'm talking about.

[0]: https://en.wikipedia.org/wiki/Elegant_variation


> * Nested parentheses. (In theory, you could use any level needed (provided that they make sense), but in practice, they're scoffed at.) It is similar with quotes, for the same reasons: limited readability - although here you can at least juggle with single, double and French quotes, depending on the language and style guide used.

Any level of nested parenthesis is fine (is what I think at least (as someone who's written a bit of Lisp of Lisp (or Racket and Clojure specifically) (because you get kinda used to it (when in a way they just become invisible (or unnoticeable))))). But nested quotations is also an issue with backtick (`) when doing command substitution in the shell, and we have to resort to $() to fix it. There's not really a $()-like syntax for written English though.


There's not really a $()-like syntax for written English though.

There sort of is, if you use "smart" quotes. “Hard to find on a phone keyboard though, and not that distinct.”


Parsing nested parens requires stack, which humans are vary bad at handling. So writing in this style will be harder to read compared to linear style that people are used to. There are topics inherently nonlinear, but they are often linearized or directly drawn in 2d space.


Also most languages already support nested clause structures without the use of parentheses.


They do, but there's a limit to how much recursive embedding the average reader can cope with, especially if the writer embeds a lot of modifying clauses in the middle of the main clause so the reader is left waiting for information they need to make sense of the sentence as a whole.


Absolutely agree. It's probably the biggest disconnect I feel between how ideas are structured in my head and how to best communicate them in written format.

Parentheticals in English sit in this sort of midpoint between a modifying clause for necessary elaboration and a footnote/end-note for optional detail, and lots of people including me use parentheses where "inlining" the parentheses or moving them to a footnote would be more appropriate. That's in addition to conventional uses of parentheses, like to expand abbreviations.


Is "vary bad" a clever pun, or just an amusing typo?


Regarding repetitons: the aversion to repeated words can create some amusingly inelegant sentences. These have become known as "knobbly monsters", after a particularly egregious example where a writer needed a synonym for "alligators". British tabloids elevated the practice to self-parody, and they are prominent among the examples recorded by @knobblymonsters: https://twitter.com/knobblymonsters .


Absolutely. My comment is only about whether or not the second period is logically redundant.


This is the kind of thing that really annoy programmers (because in programming languages, you can never omit things like this), but everyone else is perfectly fine with.

As a programmer though, it certainly annoys me!


I feel the same way, but I think programmers should hold their feelings and just pick up an APA, MLA or Chicago style guide or whatever the equivalent is if you're writing in British English and embrace the rules. Or if you're publishing, let an editor fix your stuff.

Every effort to make a logically consistent "engineered" language has so far failed by any reasonable measure, notably, Esperanto. Others have tried and had even less success than Esperanto. For things to even change in language usage there really has to be a pain point or some kind of trauma or isolation. Mere aspiration for aesthetics or logical consistency isn't enough. How did the American English come to be anyway? A bunch of people got on boats and went to a remote wild continent and stayed there. Forever.


This is sound, pragmatic advice. But it doesn't mean we have to like it.


> everyone else is perfectly fine with

My newspaper will shorten "United States" to "U.S.", and not add an extra period at the end of a sentence. (Not sure if the spaces are different widths.) When the next word would naturally start with a capital letter it can be difficult to tell whether the sentence ends after the abbreviation, sometimes making for garden-path sentences.

>> There have been some rumblings inside the U.S. Supreme Court justice Stephen Breyer said in an interview...

Though I am a programmer, so maybe it is just us...


In fact in some programming languages you can (or should) omit things like this. E.g. pascal does not require or explicitly forbids semicolon before end keyword.


See also automatic semicolon insertion in JavaScript, and conditional permission to omit closing <p> tags in HTML.

https://developer.mozilla.org/en-US/docs/Web/JavaScript/Refe...

https://html.spec.whatwg.org/multipage/grouping-content.html...


Pascal is fully self-consistent here: in it, semicolon is a statement separator, not a statement terminator. Since the statement before "end" is the last one in the block, there's no statement to separate it from with a trailing semicolon, and so it's a syntax error.


Pascal permits a null production for the statement grammar.

    simple-statement = 
              empty-statement 
            | assignment-statement
            | procedure-statement
            | goto-statement .

    empty-statement = .
from http://www.pascal-central.com/iso7185.html#6.8%20Statements (ISO 7185 Pascal)

In practice the separator-vs-terminator distinction mostly shows up in if-statements:

  if foo then
     bar
  else
     baz;
I believe that anywhere else in a statement context, you can put in as many extra ';' as you like and it won't make a difference. And of course you can drop the final ';' before 'end' or 'until'.


It's not a syntax error.


Correct - the redundant semi-colon denotes an empty statement.


In those day when moving cursors with full screen editor (which is Not vi or Emacs), how many hours of my life wasted on this deletion of semicolons. Sigh.

Still I actually struggled each time this given not exactly sure I am in English or American camp.


It is redundant because the punctuation is mainly there to indicate pausing or transitioning, but the quotes already indicate a change in voice, which is also a pause/transition.

To that end, I am struggling to think of a time where the punctuation at the end of a quote really mattered. Even exclamations are less necessary, with the support of the surrounding sentance. Consider; they screamed "stop" to get the attention of everyone. And, "stop!" they screamed. In both, the change of voice to the quoted word is about the same.


Yes; so the period inside the quotes is redundant, and that one should be removed.

The outside sentence can continue after the quote; the terminating period outside of the quote is necessary if the sentence is ending there.

We know that the quote has ended with the closing quote, and so we can agree on the convention that there is an implied period there, if the quoted material is a complete clause with subject and verb. If it needs some other punctuation like a question mark, then that is explicit.

> He asked me "what time is it?".

The ? ends the question, the . ends the sentence that contains the embedded question.

The enclosing sentence can continue:

> He asked me "what time is it?" but hurriedly walked away as I glanced at my watch.

Now you positively, definitely cannot remove the period after "watch". Why would you remove it if the words "but ... watch" are deleted?

This stuff is simply too important to leave in the hands of people who have never written a compiler.


Apologies on the long delay. I forgot I had posted. :(

I meant my point mostly tongue in cheek. Redundancy is not necessarily bad.

I do have trouble reading your sentence, though. The change of voice in a declarative sentence to interrogative without a stop, is kind of jarring. In that one, I'd leave off the quote and question mark, I think.

That all said, I think I mainly agree that it is all guidelines and mostly driven by reasons we no longer remember or care about. The rule I remember hearing was that the smaller character of the period was preferred inside the larger character, as the period was prone to break if done in the other way.


I think, if we want to leave out the question mark, we have to switch to an indirect quote (reported speech):

> He asked me what time it is, and ....

Now we have an embedded wh-clause, not an embedded quote. Embedded wh-clauses are not questions.

(One piece of evidence: subject and auxiliary are not inverted, appearing as it is rather than is it.)


What benefit does the reader draw from being informed “this quoted phrase is a complete sentence”?

It’s not futile to reconsider conventions, no matter that demands may go ignored.


Floatingatoll said "It's not futile to reconsider conventions". This shows that's not a complete sentence, suggesting that you added a caveat or other clause.

Also, your preceding paragraph highlights another issue: what if the sentence being quoted has different punctuation than the sentence it's placed in? You wrote a sentence that should end in a question mark, but the quote should end with a period.


Good illustration, but also brings into scope the [sadly disappearing?] ethical requirement of not misleading with abbreviated or out-of-context quotations.


So, when a quoted sentence fragment terminates a sentence, the period should be outside?

> Blah blah "Quoted sentence in entirety."

versus

> Blah blah "Quoted sentence fragment".


Exactly, that's the rule in British style. In American style, the period goes inside in both cases.


> Without turning the page, there would be no way of knowing whether the quoting sentence is complete.

and he said: "Mary, let's go

--- page ---

fishing".


Supposing the sentence weren't complete on the first page but the quoted one were, is this what you would expect to see?

"...and also look ugly.",

In that case, the convention of placing just a comma inside the quotation marks if the quoted sentence were complete but the quoting sentence were not seems much cleaner.


Yes that's exactly what I'd expect to see.

    He said "I came in with the tide.", but she wasn't listening.
(which is correct according to my expensive British education) is very preferable to

    He said "I came in with the tide.," but she wasn't listening.
which has a full stop and comma next to each other and looks hideous.

However, I do personally think the full stop inside the quotes is redundant and would remove it, unless it made a real difference to the sentence meaning.


> I don't see why it is redundant

Can you construct a pair of sentences where the potential ambiguity would be meaningful and not trivially resolved with context?


If a quote is incomplete and a meaningful part is omitted there are other typographical conventions to indicate that, like a double period .. for example


>If a quote is incomplete and a meaningful part is omitted there are other typographical conventions to indicate that, like a double period .. for example

Yes, but what I'm talking about is the reverse situation. How would you know whether or not the quoting (outer) sentence is complete? You cannot logically infer that from the quoted (inner) sentence, which is why I'm saying that the second period is not redudant.


you would assume its complete in the absence of a double period. For instance, he said "some random shit". The quote in the previous sentence is a complete one. But then he said " some other ..". An incomplete quote. I personally think we should include the full stop though as it conveys more information and is less cognitive overhead, "it's obviously not redundant.". Even considering that we tend to treat punctuation as breaking/pausing I still think its appropriate as the quote ended AND its containing sentence.


>you would assume its complete in the absence of a double period

That is true for the inner sentence, but again, I'm talking about the outer sentence. It needs its own terminator to make it unambiguously clear that it is complete.


I've never seen a double period used to mean anything. An ellipsis is a specific mark made by three periods. Or is this a Britishism?


It’s not used for anything. I can’t imagine where anyone got the idea that it is.


Terminating the quoted sentence with a period is redundant, since the closing quotation mark also terminates it.


No, because a quote can contain a sub-sentence. The period adds meaning to the sentence by adding finality.

"You are going" has a very different meaning from "You are going.", because the first implies there is something `you` are going to do. Whilst the second implies you are being ordered to leave.


In that case, not capitalizing the word immediately following the quote will work as an indicator that the period is not part of the parent. This doesn't work in the case that the word following is a proper noun. But that's an edge case on something that doesn't really matter.

I agree that maintaining punctuation of the quote is important for context. But outside of the quote, it practically never matters if there is a period or not after the quote. The reader may choose.

> He looks up, "This is it." Bob says, gazing at the sky.


I agree with that, assuming there would be some indication if the quoted sentence wasn't complete.

I'm just saying that the outer/quoting sentence needs its own punctuation regardless of what's going on in between those quotation marks.


Then you have no signifier for whether the quote is partial or not. Did you quote part of something someone said or the full sentence?


So would you suggest that it would make sense to never have punctuation before a quotation mark?


Only when necessary.

E.g.: '«Are you joking?!», he said. The other replied, «I'm not joking at all». They went on'.


Isn’t it a rule that if you quote a complete sentence it must end the quoting sentence?


Punctuation has a braking effect on the reader.

, ; : .

If you're at a full stop, the second . feels superfluous.


Then again, ellipsis marks are also a thing.


Did you really say, "I tend to agree with the author?"

See the problem there? The quoted sentence is not a question, but the outer sentence is; why, then, is the question mark inside the quotation marks? Double punctuation would solve that ambiguity, and would not be redundant.

Alternatively, and I'm not sure which is correct, if the question mark is outside the quotation marks, was the quote a complete sentence?


The question mark is supposed to go outside the quote if it is not part of the quote, FYI. Same with exclamation points. Internalizing only happens with periods and commas.


That's what it looked like to me, too, but isn't it telling that we can't treat punctuation consistently?


We can treat it consistently, it just takes a full book-length style guide to do so. Exceptions of exceptions are a frustrating form of consistency, but they are consistent nevertheless.


In any colloquial sense, the existence of exceptions means there is inconsistency.


Also awkward:

Did you ask, "Is this your dog?"?


To me, seeing that punctuation is soothing. It's finding another person who I know thinks the same way I do, if only about this one, pedantic thing.


I do wonder if that kind of thing is awkward just because we're not used to it...


What would be the proper punctuation if you rearranged the sentence slightly?

Dr Johnson kicked a large rock and said, “I refute it thus.”, as his foot rebounded.

Is the period redundant? Or the comma after it? None of them? This looks almost as ugly as two periods.


The period before the quotation mark is redundant, because the quotation mark implies it (in this case).


To me it looks wrong to omit a period in a famous short one-sentence statement. To stay within American style, I would rearrange the sentence to end with the quote, "I refute it thus."


I like the 'more logical still' style.

The structure of the whole looks incomplete and unbalanced without the second full-stop, which I find ugly. I guess that I care more about logical structure (or my idea of logical structure), than typographical utility.


For simplicity, every sentence should end with a full stop (period). Thus:

    How are you today?.

    This is important!.
Which clearly demonstrates that ! and ? can appear in the middle of a sentence, just as ; and , can:

   I want to call to your attention this very important!! point: always checks the return value of your system calls.
I do draw the line at introducing scoping operators though it would simplify the grammar.


The dot below ? and ! is the full stop.

Thus implying the existence of non-dotted versions for mid-sentence usage.


Good point -- I shoulda known as IPv8 addresses use ., !, and ? to separate address parts when using "dotted" notation.

> Thus implying the existence of non-dotted versions for mid-sentence usage.

Or would they have commas as is used when the : is degrades into the less definitive ;?


Do you mean, like this: ““I refute it thus.”, he said.”? It does get ugly.


Redundant and ugly are very subjective qualities in this case.


Once we're accepting that appearance is part of our decision, the American one has just as much appeal as the British one.


You mean, "typograhicl redunat".




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: