April 9, 2008 at 22:42All (or almost all) of the gap
As follows from the title, the article will focus on an integral part of any Russian-language (and not only) text - on the gap. We will cover the history of the gap, the types of spaces, the issues of using spaces in web typography.
Generally speaking, a space is any empty space in handwritten, printed, or displayed on any other medium text. So the spaces are different:
There is also a good article by Anton Bizyaev about gaps and their history “There were no gaps at the beginning” , which was published in 1997 in the journal “Publish”.
In short, the gap appeared quite late, in those scripts where the lack of differentiation of words led to difficulty in reading (the so-called consonant letter, where only consonant sounds are recorded). However, in Greek and Latin, in which vowels were recorded, the use of a space was lost. Paul Sanger connects this with the fact that reading was done out loud, which simplified the differentiation of words in the perception of the text.
Again, the space began to be used around the 7th – 9th centuries. n e., and this tradition came from Ireland, where the scribes and readers had the mother tongue of Old Irish, and religious literature was recorded in Latin. Apparently, for this reason, the monks had difficulty reading aloud. It is believed that the appearance of a gap is closely related to the gradual transition from reading aloud to reading to oneself. Examples of books in Latin with interword spaces are monuments of British literature: the Gospel of Darrow (7th century) and the Book of Kells (8th — 9th centuries).
In the Glagolitic and Cyrillic letters, there was no space either, and in the usual sense, it has been used only since the 17th century.
Before mankind invented the typeface, there was no special classification of inter-word spaces - scribes put spaces by eye and set them. Let me remind you (we wrote about this in the article “Turning the Width Off” ) that the manuscript and xylography are related to the methods of creating texts without the mobility of letters. Naturally, gaps could be obtained in different widths, since the gaps were made manually.
When the mobility of the letters appeared (and this happened with the advent of typesetting), questions arose accordingly - but how to put spaces so that the width is respected?
The manual dialing technology is such that the typed line is completely clamped in the layout and in the galley, and, accordingly, should have a width almost exactly equal to the width of the strip (for more details on manual dialing technology, see M.V.Shulmeister’s book of the same name ).
When manually typing, a line was drawn from letters (bars, on the end of which were made convex mirror copies of letters printed on paper), and word spaces were created using the so-called spacings - bars of various thicknesses, which have no printing surface on the end. It looks something like this. Spots for each font size, of course, produced their own, and had different widths. For example, for a font size of 10 points (the standard size for most text editions) were produced spations with a width of 10, 5, 4, 3, 2, and 1 point.
Spots in the size of the size were called skeletal or round. Spots in half skittle were called semicircular or semicircular. There is also the name “thin spacing”, which means spations with a thickness of 1-2 points for a font size of 8-12 points. That is, for a font size of 10 points, the thin spacing is usually 2 points (respectively, 1⁄5 of the size). However, due to the lack of an accurate definition of thin spacing, the manuals of the publisher, editor, and typesetter usually do not refer to a beat to a thin spacing, but to beat by so many points (assuming that the size of the font is 10 points).
Thus, it must be understood that, depending on the size of the font, the proportion of round spatia (third, quarter, etc.) can have different widths in points, and vice versa.
So, having figured out what round and semicircular spations are, we will move on to the width of the actual word gap adopted in the Russian set.
Schulmeister writes (p. 94) that when typing a line, semicircular is put between the words. When a line is typed to the end, in most cases its width is either less than or greater than the width of the dial strip. Therefore, the typesetter has to change the width of the spaces, reducing it to at least 1⁄4 round and increasing to a maximum of 3⁄4 round (respectively, when typing 10 points between words can vary from 3 to 7 points). Naturally, there are nuances that depend on the format of the publication, but we will not touch on them.
However, Schulmeister makes a reservation that the interword space in a semicircular space is too large, and the use of a standard space in 1⁄3 round space is both more economical in terms of paper consumption and often more beautiful. Also, using a word space in a semicircular space is not recommended for narrow fonts.
With the advent of line-casting machines, spaces began to become uniform in width within one line, and the width of the inter-word space began to vary about 1⁄3 round.
When switching to computer typesetting systems, a transition was made from specifying the width of the spacing in points to indicating the width of the spacing in fractions of a round, since the fonts began to be easily scaled to any size, and white-space elements should remain proportional to the size of the font.
Unicode provides the following characters for spaces in Western typography.
All other whitespace characters have a fixed width and do not stretch when you turn off lines in width. However, according to the Unicode line break algorithm , they should all be treated as a line break point.
Since the width of the word space is fixed in the font and changes automatically when turned off in width, the use of other white space characters as word words is justified only when typing prints, and only if there is a deep understanding of why this is done.
In the usual layout for the web, to separate words, it is enough to use regular and inextricable interword spaces.
At the same time, according to the rules of Russian-language typography, thin espionage should be used in a number of places (more precisely, reference books say about two-point espion, but we will use the term “thin espion” as the most appropriate both from the point of view of established terminology and from the point of view of the appearance of the line when typing).
The basic rules for using spaces will be described below, but in general, we recommend the following principle for use in layout for the web.
When preparing HTML documents for publication on the Internet, only space, non-breaking space, and thin spacing should be used as white space elements. In the event that the author suggests that the page should be viewed using agents that incorrectly process the character, then instead of thin spacing, a regular or inextricable space should be used.
Using only a thin espionia from the whole variety of white space elements allows, firstly, to preserve a harmonious appearance of the typed text, and secondly, not to overload the author of the publication with various rules for using spations of different fractional widths.
In preparing the material of the article, we conducted a kind of experiment on a specially prepared page. Yandex and Google deal with non-standard characters well, replacing all non-standard whitespace elements with regular ones when searching (we believe that this is the correct behavior). That is, they do not make a difference between the texts “two words”, “two words”, “two words”, etc.
As it turned out, rendering non-standard whitespace elements works very badly in browsers. Normally, only Firefox 3.0 on Windows XP and * nix, MSIE 7.0 and Safari on Windows XP can cope with the task. There is no data on MSIE 8.0, but most likely, it is also in order.
In the event that the use of such browsers is likely, we recommend using regular or inextricable interword spaces instead of thin spacing.
As described above, according to Unicode recommendations, thin spacing is a space where line breaks are possible. In cases where the rules require setting a thin spacing and prohibiting line breaks (for example, between digits when dialing a number), it is necessary to use a type construct
Next, we will describe those rules for the placement of spaces that are most often, according to our observations, violated during typesetting. More detailed information about the rules for typing can be found, for example, in the “Handbook of the publisher and author” A. E. Milchin and L. K. Cheltsova.
Generally speaking, a space is any empty space in handwritten, printed, or displayed on any other medium text. So the spaces are different:
- trigger (large vertical gaps in the front page of the publication) and trailing spaces of the strip,
- paragraph indents and trailing spaces of a paragraph,
- line spacing (between lines of text),
- interword spaces (between words on the same line),
- inter-letter spaces (between letters in a word).
History of interword space
The interword gap is a comparatively later invention in the history of human thought. The history of the gap is deeply set forth in Paul Saenger's book “Space between Words: The Origins of Silent Reading”, as well as, somewhat less deeply, in Johannes Friedrich's book “The History of Writing”.There is also a good article by Anton Bizyaev about gaps and their history “There were no gaps at the beginning” , which was published in 1997 in the journal “Publish”.
In short, the gap appeared quite late, in those scripts where the lack of differentiation of words led to difficulty in reading (the so-called consonant letter, where only consonant sounds are recorded). However, in Greek and Latin, in which vowels were recorded, the use of a space was lost. Paul Sanger connects this with the fact that reading was done out loud, which simplified the differentiation of words in the perception of the text.
Again, the space began to be used around the 7th – 9th centuries. n e., and this tradition came from Ireland, where the scribes and readers had the mother tongue of Old Irish, and religious literature was recorded in Latin. Apparently, for this reason, the monks had difficulty reading aloud. It is believed that the appearance of a gap is closely related to the gradual transition from reading aloud to reading to oneself. Examples of books in Latin with interword spaces are monuments of British literature: the Gospel of Darrow (7th century) and the Book of Kells (8th — 9th centuries).
In the Glagolitic and Cyrillic letters, there was no space either, and in the usual sense, it has been used only since the 17th century.
Before mankind invented the typeface, there was no special classification of inter-word spaces - scribes put spaces by eye and set them. Let me remind you (we wrote about this in the article “Turning the Width Off” ) that the manuscript and xylography are related to the methods of creating texts without the mobility of letters. Naturally, gaps could be obtained in different widths, since the gaps were made manually.
Spaces in manual dialing
When the mobility of the letters appeared (and this happened with the advent of typesetting), questions arose accordingly - but how to put spaces so that the width is respected?
The manual dialing technology is such that the typed line is completely clamped in the layout and in the galley, and, accordingly, should have a width almost exactly equal to the width of the strip (for more details on manual dialing technology, see M.V.Shulmeister’s book of the same name ).
When manually typing, a line was drawn from letters (bars, on the end of which were made convex mirror copies of letters printed on paper), and word spaces were created using the so-called spacings - bars of various thicknesses, which have no printing surface on the end. It looks something like this. Spots for each font size, of course, produced their own, and had different widths. For example, for a font size of 10 points (the standard size for most text editions) were produced spations with a width of 10, 5, 4, 3, 2, and 1 point.
Spots in the size of the size were called skeletal or round. Spots in half skittle were called semicircular or semicircular. There is also the name “thin spacing”, which means spations with a thickness of 1-2 points for a font size of 8-12 points. That is, for a font size of 10 points, the thin spacing is usually 2 points (respectively, 1⁄5 of the size). However, due to the lack of an accurate definition of thin spacing, the manuals of the publisher, editor, and typesetter usually do not refer to a beat to a thin spacing, but to beat by so many points (assuming that the size of the font is 10 points).
Thus, it must be understood that, depending on the size of the font, the proportion of round spatia (third, quarter, etc.) can have different widths in points, and vice versa.
Traditional interword space width
So, having figured out what round and semicircular spations are, we will move on to the width of the actual word gap adopted in the Russian set.
Schulmeister writes (p. 94) that when typing a line, semicircular is put between the words. When a line is typed to the end, in most cases its width is either less than or greater than the width of the dial strip. Therefore, the typesetter has to change the width of the spaces, reducing it to at least 1⁄4 round and increasing to a maximum of 3⁄4 round (respectively, when typing 10 points between words can vary from 3 to 7 points). Naturally, there are nuances that depend on the format of the publication, but we will not touch on them.
However, Schulmeister makes a reservation that the interword space in a semicircular space is too large, and the use of a standard space in 1⁄3 round space is both more economical in terms of paper consumption and often more beautiful. Also, using a word space in a semicircular space is not recommended for narrow fonts.
With the advent of line-casting machines, spaces began to become uniform in width within one line, and the width of the inter-word space began to vary about 1⁄3 round.
Computer typography and web typography
Currently, we are limited by the capabilities of the fonts used, and, of course, by the character set in Unicode. It must be remembered that not all fonts contain most whitespace Unicode characters.When switching to computer typesetting systems, a transition was made from specifying the width of the spacing in points to indicating the width of the spacing in fractions of a round, since the fonts began to be easily scaled to any size, and white-space elements should remain proportional to the size of the font.
Unicode space characters
Unicode provides the following characters for spaces in Western typography.
- Interword space , U + 0020, - width from 1⁄5 to 1⁄2 round depending on the font. For medium fonts, the interword space has a width of the order of 1⁄4 round (for example, Times New Roman has just such a space), for the wide fonts - of the order of 1⁄3 of the round (Microsoft Verdana - 0.35 round, Microsoft Tahoma - 0.31 round).
- Non-breaking inter-word space , U + 00A0, - has the same width as a regular inter-word space, but line break is prohibited in the place of non-breaking space.
All other whitespace characters have a fixed width and do not stretch when you turn off lines in width. However, according to the Unicode line break algorithm , they should all be treated as a line break point.
- Round spice , U + 2003, - as was said, has a width equal to the size of the size. Also called Em Space, perhaps because the letter “M” in some old fonts had such a width. However, now this is far from being carried out everywhere, and therefore the assertion that Em Space always has the width of the letter “M” is a misconception.
- Semicircular spatia , U + 2002, is half round. Also called En Space, perhaps because the letter “N” in some old fonts had such a width. However, now this is far from being carried out everywhere, and therefore the assertion that En Space always has the width of the letter “N” is a misconception.
- The third spation , U + 2004, is a third of the round. In English is called Three-per-Em Space.
- Fourth spation , U + 2005, - a quarter round. In English called Four-per-Em Space.
- One Sixth Round , U + 2006,. In English is called Six-per-Em Space.
- Thin espionage , U + 2009, - usually has a width of 1⁄5 round (less often - 1⁄6). Generally speaking, its width depends on the typing language and font manufacturer, and in cyrillic fonts, thin spacing usually has a width of 1⁄5 round. This proportion in proportion exactly corresponds to a two-point spacing when typing with a size of 10 points. In English called Thin Space.
- Hairy Spatius , U + 200A, is the narrowest, with a width of about 1⁄10-1⁄16 round. Such a spacing in proportions approximately corresponds to a single-point spacing when typing with a size of 10 points or looks even narrower.
Using various spaces
Since the width of the word space is fixed in the font and changes automatically when turned off in width, the use of other white space characters as word words is justified only when typing prints, and only if there is a deep understanding of why this is done.
In the usual layout for the web, to separate words, it is enough to use regular and inextricable interword spaces.
At the same time, according to the rules of Russian-language typography, thin espionage should be used in a number of places (more precisely, reference books say about two-point espion, but we will use the term “thin espion” as the most appropriate both from the point of view of established terminology and from the point of view of the appearance of the line when typing).
The basic rules for using spaces will be described below, but in general, we recommend the following principle for use in layout for the web.
When preparing HTML documents for publication on the Internet, only space, non-breaking space, and thin spacing should be used as white space elements. In the event that the author suggests that the page should be viewed using agents that incorrectly process the character, then instead of thin spacing, a regular or inextricable space should be used.
Using only a thin espionia from the whole variety of white space elements allows, firstly, to preserve a harmonious appearance of the typed text, and secondly, not to overload the author of the publication with various rules for using spations of different fractional widths.
Space handling by browsers and search engines
In preparing the material of the article, we conducted a kind of experiment on a specially prepared page. Yandex and Google deal with non-standard characters well, replacing all non-standard whitespace elements with regular ones when searching (we believe that this is the correct behavior). That is, they do not make a difference between the texts “two words”, “two words”, “two words”, etc.
As it turned out, rendering non-standard whitespace elements works very badly in browsers. Normally, only Firefox 3.0 on Windows XP and * nix, MSIE 7.0 and Safari on Windows XP can cope with the task. There is no data on MSIE 8.0, but most likely, it is also in order.
- Firefox prior to version 3.0 does not break the line at all in places of non-standard spaces. In this case, the width of the spaces is displayed correctly.
- Opera 9.26 and 9.50, FireFox 3.0 for Mac, Safari for Mac carry the line, but all the spaces are the same width.
- MSIE 5.5 and 6.0 for Windows 2000 instead of spaces put squares (perhaps the corresponding characters are simply absent in the system font).
Basic rules for using spaces
So, we emphasize once again that in all the rules listed below, thin spacing is used only when the author discards the risk of a visitor using browsers that incorrectly display thin spacing. These include some browsers in * nix (perhaps this is due to the built-in fonts), MSIE version 6.0 and earlier, browsers for Mac (they can be neglected, since the rendering error is only the width of the spits), maybe some browsers for mobile phones and PDA.In the event that the use of such browsers is likely, we recommend using regular or inextricable interword spaces instead of thin spacing.
As described above, according to Unicode recommendations, thin spacing is a space where line breaks are possible. In cases where the rules require setting a thin spacing and prohibiting line breaks (for example, between digits when dialing a number), it is necessary to use a type construct
250 000. The HTML element nobris proprietary and is not allowed to be used. Next, we will describe those rules for the placement of spaces that are most often, according to our observations, violated during typesetting. More detailed information about the rules for typing can be found, for example, in the “Handbook of the publisher and author” A. E. Milchin and L. K. Cheltsova.
Abbreviations and Symbols
- В сокращениях «и так далее», «и тому подобное», «так как», «то есть», «и другие», «до нашей эры», «южной широты» и подобных все элементы сокращения отделяются неразрывным пробелом.
и т. д. — и т. д.
и т. п. — и т. п.
т. к. — т. к.
т. е. — т. е.
и др. — и др.
до н. э. — до н. э.
ю. ш. — ю. ш. - Initials are beaten off from each other and from a surname by an inextricable space.
A.S. Pushkin - A.S. Pushkin
J.R. R. Tolkien - J.R. R. Tolkien
It is also possible to discard the initials from each other and from the surname following it with a thin spacing, however, transferring the initials or surname to the next line prohibited. Regardless of the choice of style of beating initials, it is necessary to adhere to the unity of style throughout the entire document or site.
V.V. Putin - V.V. Putin
V. Putin - V. Putin
Putin V.V. - Putin V.V.
Putin V. - Putin V. - The abbreviated word is beaten off in the name of its own by an inextricable space.
st. Schorsa - st. Schorsa
Moscow - Moscow
Metro them. Lenin - Metro them. Lenin - The number and the corresponding counting word are beaten off by an inextricable space.
12 billion rubles - 12 billion rubles
hl IV - ch. IV
pp. 3-6 - pp. 3–6
fig. 42 - fig. 42
XX century - XX century
1941-1945 - 1941-1945
chamber number 6 - chamber number 6
§ 22 - § 22
25% - 25%
97.5? - 97.5?
16 ¢ - 16 ¢. - The number and the corresponding unit of measurement (except for the signs of degree, minute and second) are beaten off by a fine spacing, line break is prohibited.
400 m - 400 m
100 t - 100 t
451 ° F - 451 ° F
but 59 °, 57 ′, 00 ″. - The signs of degree, minute, and second are beaten off by a subtle escaping from subsequent digits.
59 ° 57 ′ 00 ″ - 59 ° 57 ′ 00 ″
Numbers and Intervals
- Fractional and integer parts of a number are not beaten by a space from a comma: 0.62, 345.5.
- The digits of a number are beaten off by a thin spacing, except for dates, numbers (for example, documents), designations of machines and mechanisms.
25 563.42 - 25 563.42
1 652 - 1 652
1 298 300 - 1 298 300
but 1999, GOST 20283, int. No. 982364 - With the numerical designation of the intervals, the dash does not deflect from the boundaries of the interval.
50-100 m - 50-100 m
1,500-2,000 - 1,500-2,000
1,5-2,000 - 1,5-2,000
15-20% - 15-20% - The unary signs plus, minus and plus or minus are not beaten off from the number following it: +20 ° C, −42, ± 0.1.
- Binary signs of mathematical operations and relations are beaten off on both sides by a delicate spacing.
2 + 3 = 5 - 2 + 3 = 5
Punctuation marks
- The dot, comma, colon, question and exclamation points, semicolons do not break with a space from the previous word, and break with a space from the following: Ha, ha. Ha? Ha!
- The ellipsis does not stray from the previous word, if it is at the end of the sentence or part of the sentence, and from the next, if it is at the beginning of the sentence: Wow ... What? …Nothing.
- The quotation marks do not break off with spaces from the text enclosed in them: the battleship “Potemkin”.
- Brackets are not beaten with spaces from the text enclosed in them, and are beaten with spaces outside (except for the case when the closing bracket is adjacent to the punctuation mark on the right): Text in brackets is of no interest to anyone (usually).
- A dash is beaten off from the previous word by an inextricable space, and from the next by a regular space (including if the interval is indicated in verbal rather than digital form).
Vitenka - well done!
only a cucumber of fifteen to twenty centimeters
Molotov-Ribbentrop Pact is suitable for us . - If two numbers in verbal form do not form an interval, but mean “either one number or another”, then a hyphen is put between them, which does not break with spaces: I drank two or three glasses.
Unwanted transfers
- Short words and unions ( a , and , but , me , you , and so on) are better to beat off from the next word with an inextricable space, since the short word hanging at the end of the line impairs readability. In particular it is very desirable to prevent the transfer line between the particle not and following verb.
- Particles of the same , would , if it is desirable to beat the previous word non-breaking spaces: the same, I would say, I thought.
- It is advisable not to tear off the prepositions at the beginning of the sentence from the words following them. (even longer than one and two letter)
