Slashdot Mirror
Are Two Spaces After a Period Better Than One? (arstechnica.com)
Researchers at Skidmore College conducted an eye-tracking experiment with 60 Skidmore students and found that two spaces at the end of a period slightly improved the processing of text during reading. Ars Technica reports the findings: Previous cognitive science research has been divided on the issue. Some research has suggested closer spacing of the beginning of a new sentence may allow a reader to capture more characters in their parafoveal vision -- the area of the retina just outside the area of focus, or fovea -- and thus start processing the information sooner (though experimental evidence of that was not very strong). Other prior research has inferred that an extra space prevents lateral interference in processing text, making it easier for the reader to identify the word in focus. But no prior research found by [study authors] Johnson, Bui, and Schmitt actually measured reader performance with each typographic scheme.
First, they divided their group of 60 research subjects by way of a keyboard task -- the subjects typed text dictated to them into a computer and were sorted into "one-spacers" (39 regularly put a single space between sentences) and "two-spacers" (21 hit that space bar twice consistently after a period). Every student subject used but a single space after each comma. Having identified subjects' proclivities, the researchers then gave them 21 paragraphs to read (including one practice paragraph) on a computer screen and tracked their eye movement as they read using an Eyelink 1000 video-based eye tracking system. [...] The "one-spacers" were, as a group, slower readers across the board (by about 10 words per minute), and they showed statistically insignificant variation across all four spacing practices. And "two-spacers" saw a three-percent increase in reading speed for paragraphs in their own favored spacing scheme. The controversial part of the study has to do with the 14 point Courier New font that the researchers presented to the students. "Courier New is a fixed-width font that resembles typewritten text -- used by hardly anyone for documents," reports Ars. "Even the APA suggests using 12 point Times Roman, a proportional-width font. Fixed-width fonts make a double-space more pronounced."
First, they divided their group of 60 research subjects by way of a keyboard task -- the subjects typed text dictated to them into a computer and were sorted into "one-spacers" (39 regularly put a single space between sentences) and "two-spacers" (21 hit that space bar twice consistently after a period). Every student subject used but a single space after each comma. Having identified subjects' proclivities, the researchers then gave them 21 paragraphs to read (including one practice paragraph) on a computer screen and tracked their eye movement as they read using an Eyelink 1000 video-based eye tracking system. [...] The "one-spacers" were, as a group, slower readers across the board (by about 10 words per minute), and they showed statistically insignificant variation across all four spacing practices. And "two-spacers" saw a three-percent increase in reading speed for paragraphs in their own favored spacing scheme. The controversial part of the study has to do with the 14 point Courier New font that the researchers presented to the students. "Courier New is a fixed-width font that resembles typewritten text -- used by hardly anyone for documents," reports Ars. "Even the APA suggests using 12 point Times Roman, a proportional-width font. Fixed-width fonts make a double-space more pronounced."
Is it any wonder that the brain is optimized for 2 spaces after a period?
There's probably a part of the brain which is totally optimized for recognizing that particular text cue by the time you have read a few thousand sentences.
It's a fixed format font- so it will be different for proportional fonts and the sites I use already display both single and multiple spaces after a period as 'about 2 spaces". They don't alter the textual data- they just alter the way the text is displayed.
There's probably a benefit to some spacing difference vs "all run together text.with nospaces." But if everyone had been reading text with 1.5 or 2.7 spaces after a period since age 3 then the test would probably have found that was the ideal spacing.
She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
Adding extra space after the period makes sense typographically. That's why it make sense to put two spaces after a period when typing on a typewriter.
Computers can add the space automatically though. If you look at the spaces after a period in any decent font, it's wider than the spaces after other letters.
"First they came for the slanderers and i said nothing."
If you have a search function that doesn't ignore whitespace, you're probably onto a loser before you even start.
Plus, why would you search for an entire sentence + follow on sentence in one search? That's just going to end in disaster whether it's a double-space, a new-line, a new paragraph or anything else.
"Ignore whitespace" is an option in your searches for a reason.
Their test subjects were 60 college students. It's reasonable to assume that most of them have never seen a typewriter in the flesh, much less used one.
So you're angry that the "old type-writer-using geezer" is reading ten words per minute more than you and you're trying to sabotage his productivity?
And millennials wonder why nobody likes them...
"But we had one certain thing about a human language, - that the words are separated with a space, with one space."
I hate to break it to you.... people (like myself) were taught to use two spaces after a full-stop for DECADES. The predecessors of ubiquitous computing were all taught like that as the only reference was typewriters and which were often taught to double-space.
This is not a "now we'll have a problem". This is a "You've already had this problem for a very long time and it's never been standardised, so don't make bad assumptions".
You are much more likely to find existing users using double-space-after-period than you are younger users.
Fact is, there is no such standard. What standards they were stayed the same from typewriters to PCs, but then changed to this "one-space" system, and now people are arguing over "one-space vs two-space" again. So you have to code to account for all situations anyway.
Sort is more problematic, sure, but again ignoring whitespace in sort is incredibly easily (and actually beneficial... did you enter the book title as "whitehouse" or "white house"? Surely you want those listed close to each other).
This was never a given. I was EXPLICITLY taught, less than 30 years ago, to double-space after periods. It's a habit you'll find throughout this post (which I'm writing at 100+ WPM). Maybe the new generations weren't but you can't just assume that.
Don't even get me started on spaces (and punctuation) when they are near / inside the end of quotes, semi-colons, etc.
Stop making bad coding assumptions where they relate to human-entered or human-visible data (which INCLUDES search criteria, and the original entered data).
"... in any decent font, [the space after a period] is wider than the spaces after ... letters."
Exactly. The "researchers" at Skidmore ignored the fact that fonts have already solved the problem. In the future, I suggest that they skid less.
So, your comment got me wondering when the typewriter was invented. Turns out it was 1868. That's about 400 years after the printing press, so there's a lot of history to look at from *before* the typewriter existed.
And here's an article that does just that:
https://web.archive.org/web/20...
I agree with this guy. It's an aesthetic preference, so there's no right or wrong answer. I tend to prefer additional space after the end of a sentence, because it more easily allows me to see the logical break that should be represented by that sentence end. Since computers and displays today are capable of micro-adjustments to character spacing, and they also can tell where sentences end (unlike a typewriter), it's irrelevant how many spaces there are after a period in the source - the text can be (or should be able to be) displayed with my preferred spacing.
- The Sigless Wonder
Typing two spaces at the end of a sentence indicates a between-sentences gap, as opposed to a between-words gap. That may or may not be the same amount of space. If you're using a proportional font, then you're relying on the computer to handle the spacing, and this should be no different. Perhaps two spaces should kern together to be the equivalent of 1.2 spaces, and that sort of rule can be handled by the font.
The problem is that determining the difference between the end of a sentence and a period that just ends an abbreviation is quite difficult. That's something that requires natural language analysis, not something simple like kerning that is part of the design of proportional fonts.
So everyone should continue to learn to type with two spaces after a period. We know it's superior for fixed-space fonts. Computers should be taught to do the right thing for proportional fonts. If they don't, file a bug report.
Fonts did not solve this problem. Do you select a different space after a sentence or do you just press the space bar? The characters in a font do not know where they are. If you are lucky some software will fix this when the text is presented. Otherwise, it will not.