Double-spacing after periods

whoisdan's picture

Jonathan Parker's question about "middle-dots" got me wondering about the origin of the "two spaces after a period" rule I learned in typing class, and unlearned in a design class. What's the deal with that? Where'd it originate? Why do people still do it?

William Berkson's picture

I think I read that it started in the 19th century with the 'modern' faces, which were less readable in small sizes. The extra space helped.

It became standard with typewriters for the same reason I am guessing. With a monospaced font, the extra space made the ends of sentences clearer.

It persists because people don't know better and treat the computer like a typewriter - or have a misguided notion that it is 'proper'.

Nick Shinn's picture

The extra space, on a typewriter, is typographically functional in that it reduces the incidence of "divorcees". That's the situation where you get the first word of a sentence, often "A" or "The" at the end of a line -- which is something that I manually fix in rag right text, and in Web text by using a "nbsp" after the article when it is followed by a capitalized name (if I remember).

However, a better explanation is that the widths of the points in a monospaced font are hugely wider, relatively speaking, than in a proportional font. So that the character sequence "comma-space-lowercase" within a sentence really splits it, visually, and is much closer in status, as a visual marker, to the sequence "period-space-capital" that denotes a new sentence, than in a proportional font. A double space compensates and puts back the degree of distinction.

Mark Simonson's picture

Wow, I've never heard that explanation before.

From a book called "Linotype Keyboard Operation" published by Mergenthaler Linotype in 1930:

"Sentence Ending. Between sentences drop a spaceband and a thin space. This will give you a space of approximately one em."

It doesn't say why, though. I wonder when this practice changed?

William Berkson's picture

Hmmm. I am looking at a sample from Linotype in 1928 and the space after a sentence may be slightly more than a word space, maybe. But definitely not an em space, or two or three times a word space. Since a word space varies with the font, I wouldn't think that adding a thin space, which Felici says is a quarter em, would make most after sentence spaces into a full em.

Perhaps some of the old timers here, who actually worked on Linotype machines can enlighten us.

I checked and where I read about the 19th century as the start of the practice of wide spaces after sentences was in Bringhurst.

Nick Shinn's picture

>Wow, I've never heard that explanation before.

Just a theory; not entirely convincing, but ya gotta figure there are functional/aesthetic reasons for a practice so prevalent and persistent.

Following a brief analysis of mass periodicals in the Shinn library, it appears to extend (for basic editorial text) from before the 19th century, into the 1950s.

So it looks like it was common practice for the majority of the modern era.

Perhaps the arrival of the rag right International Style, with the extreme evenness of its wordspacing, finished it off.

I wonder if it has any benefits for convenience in hand composition.

gerald_giampa's picture

Nick,

I can't see any benefit for the hand compositor by adding double spaces after a period.

http://lanstontype.com/TypeStick.html

Think about it! The more you have to set, the more you have to set? A pretty simple equation.

Not to mention, additional spaces in lengthy works will creep the text into extra pages, extra pages mean extra paper and extra paper means extra bindery, extra press runs, slower schedules, more ink. More depreciation on the machinery, missed schedules, or "overtime".

Sounds like a bad idea to me.

The physical space after a "point" (period) should be thinner to compensate for the optical whiteness above the "point". Even spacing, even reading.

Double spacing was just a bad fad. Don't believe everything your read in a Linotype book.

kentlew's picture

William, as far as I know, the term "thin space" does not have an absolute, universally accepted dimension (despite what Mr. Felici states). I've seen it just as often defined as M/5, or even M/6, as I have M/4. I haven't been able to identify if it had a specific defined value in the context of Linotype typesetting.

Also, in Linotype setting (and also hand composition, for that matter) the spaceband is not part of a font of type. Lino spacebands go in a separate holding area of the machine and get inserted through a different mechanism than the characters of a font.

The Linotype spaceband is a long, tapered piece; this taper is what allows the line to be justified and all the word spaces adjusted evenly to fill out the line. The thin top end of the taper defines your minimum word space; the thickness at the bottom end defines the maximum. I believe you could order Linotype spacebands of different minimum widths (and possibly even different tapers).

Adding a thin space (however defined) after an end-of-sentence spaceband would ensure that this word space was wider. But that statement about approximating an *em* space doesn't make sense.

-- K.

Mark Simonson's picture

I thought that was odd, too. The book is presumably set using this spaceband + thin space method and the lines are justified right and left. On somewhat loose lines it does look like about an em (all the paragraphs appear to be indented an em). It may be that it looks wider because of the space above the period.

In any case, there is no explanation that I can find in the book either about why it's a good idea to have that much space between sentences or why a spaceband + a thin space is aproximately an em.

(Perusing the book, it strikes me that the Linotype was a hideously complex machine to operate. You could actually damage the machine by setting too little space between words!)

jim_rimmer's picture

I guess I qualify as one of the oldtimers who actually worked on the Linotype machine.

Linotype aside, in metal type days it was split school on the matter of an extra space folowing a period in text. My ITU lesson books stated that it was a house style. In some shops it was prefereed in others it was not.

At one time I followed the notion that a thin space added was the thing to do, just becuase that was the style at JW Boyd & Sons. Now it jangles me to see an added space. To my eye, the white space above the period goves all the break needed. On the Monotype and in hand composition, I will drop even the word space if the next sentence starts with a cap T (on tight lines), and use only a thin space on more open spaced lines.

Mark: You could indeed damage things on the Linotype or the Intertype with tight lines. The damage was usually confined to bending the tails on the space bands when they were forced upwards to spread the line. The fewer spacebands there were in the line, the more likely the bands were to be bent. Another resultof tight lines was the mangling of the ears on the matrices, whose function was to assure alignment of the characters.

Very loose lines caused the slow buildup of typemetal on the sides of the mats at he casting point. The result was that molten metal was allowed to flow between the loosely packed matrices and "hairlines" or fins developed between each letter. The longer these mats were kept in service the worse it got.

Take a look at the columns of text in any old hotmetal-produced newspaper, and you will see the hairlines.

There was a hand tool available for Linotype machines which was used to drive out the side walls of the hairlined mats so that they would again cast cleanly without creating fins.

Jim Rimmer

William Berkson's picture

>the spaceband is not part of a font of type

Thanks, Kent, that is interesting. Was the thin space also a wedge? From your observations and those of Mark and Jim, I am concluding that usually the word spacing in linotype material was rather loose. I actually like the relatively wide word spacing in a linotype-set book I have.

It may be that with wider word spacing, the extra thin space helped a bit to mark sentence beginnings - or at least that was perhaps the concept. Jim, do you know whether during the same period the practice of the extra thin space was something also done on Monotype machines? Did the Monotype technology generally result in tighter word spacing than the 'wedge' technology in Linotype? Or is all this extra space stuff just a hold-over from the practice with 19th century 'modern types', which Bringhurst notes?

Thomas Phinney's picture

Interestingly, these days in digital type the average word space built into a font is about 1/4 em (the usual range is about .2 to .3, but .25 is about average). So defining a "thin space" as 1/4 em is sort of funny, in modern usage, seeing as the thin space by definition ought to be narrower than a word space.

I believe that the thin space in InDesign is 1/6 of an em, which would be 2/3 of a typical word space. This seems to me a reasonable definition for digital typesetting. Apparently it's also the default value for a thin space in TeX.

T

William Berkson's picture

To be fair I should quote Felici (p. 26): "The thin space is often defined as half an en space (or one quarter of an em), but its relationship to an em can usually be defined within most computer typesetting programs."

I am wondering whether there was a shift to much tighter word spacing with the advent of photo type. Does any one know about this? For example, what was the typical Monotype metal word space, or range of spaces? How was the word spacing adjusted in Monotype?

It seems like there is an interesting story here about word and sentence spaces changing as a result of changing design (Didot & Bodoni vs old style) and technology and fads.

hrant's picture

> HTML will not accept two spaces in a row

Which is fascism, not education.

hhp

William Berkson's picture

Extending Nick's analysis, we get the following on the relation between word and sentence spacing. If the word space is say 1/4 em and the period takes up say 1/6th em, then the space above a period increases the total space after a sentence to 66% over a word space alone. If however the word space is 1/2 em, and the space of the period remains the same, then the increase of the end of sentence space is only 33% over a word space. So with looser word spacing, it makes sense to add extra space between sentences, whereas with tighter spacing it does not.

gerald_giampa's picture

William

I am not Jim, and have not operated a Linotype. Monotype yes.

Anyway, in part this may answer some of your questions.

Your question.

Jim, do you know whether during the same period the practice of the extra thin space was something also done on Monotype machines? Did the Monotype technology generally result in tighter word spacing than the 'wedge' technology in Linotype?

Thin space bands eventually cause metal fins. The same is true of wide settings on a Linotype. The extra space after a period however has nothing to do with that. Or, at least, was incidental to setting technology.

Jim and I have both indicated the extra space after a period is a matter of "house style". My words were, a "bad fad". And yes, certainly Monotype machines had some bad fad artists tapping keys.

But for sure, dreadful typography could be set by hand, set by Linotype or set by Monotype. Just the same as in digital typography. Machines do not set type, people do.

Very tight spacing was possible with a Monotype, but not necessarily. Also it did not cause fins on a Monotype. Not the kind you see on the printed page.

The operator had exceptional control over setting text. Both variable and fixed spaces were available. Micro fine increments combined with the ability to unit add and unit subtract gave the operator the ability to do an exceptional setting.

The ability to unit subtract meant you could easily kern on a Monotype. Not so easily by hand setting. Hand setting was a matter of very fine hack saws, gravures, and files. Great care had to be made in that operation or characters could lean, or work upwards and become bolder. However good craftsmanship is a wonderful tool.

That particular difficulty was not a problem with Monotype kerning.

But all this depends on the knowledge and skills of the operator and the "nature of the craftsman".

But certainly, a slovenly mess can be made using any typesetting technology. And one way to be slovenly is to subscribe to adding an extra space after the period.

There is a good book called "The Finer Points of Spacing".

pstanley's picture

I wonder whether, in office typing, the practice was related to that other fetish: the addition of full points after Mr, Mrs and any abbreviation. Business correspondence is usually full of these things, and therefore scattered with full points that do not mark the end of sentences. Perhaps a little extra space was thought useful to prevent confusion where the full point was being used to do its real job?

I don't think this can be the origin of the practice (because it is obviously present in 19th century books, where my explanation would hardly apply), but it might explain and even to some extent justify its use with monospaced typewritten business correspondence.

A separate question: in earlier printing it was also customary to put a little extra space before colons and semicolons. Was this just a fad too? Is it ever justifiable? Because they rise above the baseline, I do sometimes feel that these marks are set a bit too close against their preceding text in some fonts at least at small sizes and affect legibility. I'm inclined to think it might sometimes be justifiable, since Finer Points (otherwise a great enemy of excessive space) reckons that a hair space might be justified in some such cases. I'm talking about a really fractional space, here, around 1/12 em at most, and I'm sure it depends on the previous letter. But kerning tables could presumably handle it.

hrant's picture

> in earlier printing it was also customary to put a little extra space before colons and semicolons.

Note that the French still do it.

The tricky thing with the colon is that it has to work well both in text and between two numerals (for time).

hhp

kentlew's picture

>I am concluding that usually the word spacing in linotype material was rather loose.

William, I think this is an overly broad generalization.

As Gerald points out, word spacing is far often more a matter of style than technology.

Generally, wider word spacing was more fashionable in the early part of last century. So much so that you find folks like Dowding railing against it.

That said, however, I think that today's digital typesetting is far more inclined toward atrociously loose spacing than ever before -- what with Quark's default of 110% optimum spacing (what a travesty) and H&J algorithms that quickly give up and just run a line out.

Indy may make some of this moot in time.

Although what Thomas says it largely true -- the average word space in the average font (i.e., Adobe) is around 250 -- there are a number of common Monotype fonts, for instance, with huge spaces of 278 for some inexplicable reason.

I, for one, wish that spacing in layout applications could be determined independent of the word space character built into the font -- much like in the old days.

-- K.

hrant's picture

Really? I'm all for letting users change the word spacing, but simply ignoring the "default" space that a type designer has (nominally) so carefully determined seems harmful.

hhp

William Berkson's picture

>overly broad generalization

Could well be - I was looking at a sample of 3 things! :-)

I was thinking that to avoid breaking the machine, and because of the fashion then, they went for looser word spacing.

Does the Quark default affect letter spacing as well? (I have InDesign)

I kind of like the big word spaces in the lino samples I am looking at, but I am here to learn.

Kent, why do you oppose the bigger spaces? And if they are to be varied at the discression of the person setting the type, what considerations would lead to a bigger or smaller word space?

I do think that the 19th century practices were a big mistake. Didot, which looks extremely elegant in display, is too dark and the serifs too fine in text. To compensate for this, I think they needed more leading and more spacing between words and hence, I am guessing, between sentences. But the whole thing still didn't work as well for text as older designs.

[IIRC, when my wife installed a French keyboard for Word, it automatically inserted a space before the colon and semicolon, and double spaced after periods!]

But I wonder whether the pendulum has swung too much the other way. Obviously you think not. If you could explain why I'd be grateful.

I have to put the Dowding book on my list also!

jim_rimmer's picture

William

I think Gerald and Kent have probably answered most of your questions about spacing on the Linotype and Monotype.

Let me add this:

There is not really a typical word space on the Monotype any more than there is on the computer in the setting of justified text matter (a book for example) since the word space spreads or pulls in at the command of the person on the keyboard. I suppose setting the H&J's on a Mac allows one to have very tight or very loose ranges of spacing.

On the Linotype, as has been mentioned, the spacebands (long, thin wedge devices) have a minimum width when they are not given the opportunity to be pushed upward by the spaceband bar to spread the line out to fill the measure. If an operator left a lot of space at the end of a line, and there were only five or six spacebands in the line, there would be a greater amount added to each word space than if there were ten spacebands and the same amount of space at the end of the line. Whatever is left over when the operator of the Lino sent the line in for casting, is spread out by the spacebands. The same sort of thing happens on the Monotype Keyboard when the operator decides that the line just keyed in is complete.

An operator wishing to set with tight word spacing on the Monotype would be conscientious in fitting the line as tightly as possible before punching in the justification codes on the tape. In either the case of the Mono or the Lino, the result is the same.

Regarding the fixed spaces available on the Linotype: in addition to the regular spacebands, there are also available narrower spacebands that will give much tighter wordspacing. I am not aware if there are any other than the two widths. In the case of large type setting on the Linotype, say 18 point to 48 pt, the operator needed to add a correct amount of fixed thin spaces to each spaceband to avoid too skinny a word space. The linotype has fixed Mut, nut and thin spaces in the magazine, available at the stroke of a key. In addition there are brass and steel thin spaces which the operator places by hand throughout the line of matrices in that have been dropped into the assembler box of the machine by the operator's keystrokes. After the line is cast, these very thin spaces are retrieved again by hand by the operator before the line is sent off to be distributed into the magazine.

The Monotype (in addition to the flexible wordspace used in justified setting) has the following fixed spaces: Em(18 unit), 12 unit, 10 unit, nut (9 unit), 8 unit, 6 unit and five unit.
By a magical manouevre by the operator it is possible to have access thinspacing as small as 2 units. This is a trick I have never been able to muster up the ambition to use. I did fall upon it accidentally once about 20 years ago, but have forgotten how it works. Too lazy to read the manual. I know Gerald Giampa is a whiz at the practice of thinspacing with the green justification key. This is a handy thing to have in the thinspacing of small cap lead-ins and in very tight lines where one needs less than a five unit space.

Summing it up: the Linotype justifies by a strightforward mechanical process of wedging out the spaces between two pre-set steel jaws (the mold vise). The Monotype counts off every unit struck by the operator as each character or space is struck. When you are at what you feel is the end of a line, you are shown how many units are left in the line for the spaces to take up un a revolving mathematical drum. In that regard, the Monotype works in exactly the same way as the computer, in that what has been keyed is computed and resolved. Or more precisely, the computer works in the same way as the Monotype.

Incidentally, there is a man in California by the name of Bruce Wasbish who has done marvelous bookwork using the Linotype. He once told me that it is possible to do fine books on the Linotype "but it is excruciating". (his words)

Sorry for this lonnnnnnnnng post.

Jim Rimmer

kakaze's picture

"There is some hope. It's interesting to note that HTML will not
accept two spaces in a row and will automatically remove the
second. Nothing on the web is double-spaced - even the stuff
that isn't part of a publication (like those I mentioned). Maybe
now, with the advent of online forums and an increasingly web-
aware population, people will start to get the idea that the
practice isn't necessary."


I notice that it eats the extra spaces, but I do it anyway.

Call me a heretic, but I prefer two spaces after periods. I read a lot, and some books seem to have NO space at all after periods. Of course, when that happens, it's a lot easier to read two sentences as one sentence, which makes you have to go back and reread it again; which can be very annoying if it happens many times during the entire book.

gerald_giampa's picture

Chris,

Of course, when that happens, it's a lot easier to read two sentences as one sentence,...

You will find this has nothing to do with the space after a period. It will have something to do with leading, but more often, how many words to a line.

kakaze's picture

"So why is one space not just right? Like it's been mentioned, the
space above the period plus one space is enough to set
sentences apart. If you believe that more space is needed, you
and I have very different notions of even color and readability."

And what I said about some books seeming to have no space at all after periods just got lost huh?

I've noticed it a lot when laying things out as well, that one space is never enough for me. I want something there that truly separates sentences from one another, and one space just doesn't cut it for me.

I guess I do have "very different notions" than you.

----

"You will find this has nothing to do with the space after a period. It will have something to do with leading, but more often, how many words to a line."

Gerald, I don't mean two sentences on separate lines, I mean like: "See Jane run. See Spot run." turning into "See Jane run See Spot run." because the period just seems to get lost.

gerald_giampa's picture

Jim,

Thank you. This morning I re-read my comments about Monotype spacing. My comments wallowed in the simplistic. I prefer your explanation, because it is one.




whoisdan's picture

Unlike Stephen, I didn't really think about why on Earth I was putting two spaces after sentences. A grad student mentioned it to me and I had no idea why I was using 2 spaces. New-sentence-indicators are there. There is a period, a space, and a capital letter all working to let you know that you're about to start a new thought. Why do you still need that extra space? I also think that in text it leaves a ton of open space, which looks pretty silly and amateur.

By the way, I'm thrilled by the feedback in this thread. It's sure developed into a whole new monster (a nice, happy one), which is educating me well beyond my initial hopes. Thanks!

kris's picture

I suppose if it looks right, it is right.

kris.

gerald_giampa's picture

Chris,

One would think that the inclusion of a period, which has meaning, an optical space from the whiteness above the period in combination with a word space, plus the upper case "S" in "See Jane run. See Spot run." would be sufficient.

I would love to have seen you in the same room as ee cummings.

I suspect your difficulties are other than what you imagine. I believe if you read Dowding's books he goes over your questions in great detail. I do not think anyone should attempt to set type without his book, "The Finer Points of Spacing", or/and Bringhurst's works. Bringhurst wrote his long after I was involved in the trade so I made do, very well with "The Finer Points of spacing."

Dowding makes clear point. Also, he illustrates point. In fact he illustrates it so well that you will find a contradiction in your suspicions.

Get the book, you will never let it far from your work place.



gerald_giampa's picture

I had the good fortune of training by an exemplary craftsman, he knew about spacing, he knew about fine books.

http://lanstontype.com/WilHudson.html

No discussion about spacing is complete without mention of "Rivers", or hanging punctuation somewhat in the margins.

kentlew's picture

Jim , great explanation of Mono and Lino justification above. I wonder, however, how much sense any written explanation can make in the absence of seeing a machine in operation, unfortunately. I don't know that I would have really followed the Mono explanation if you hadn't shown me last year.

Maybe at TypeCon this summer, we can plan a field trip to Arion, or someplace like that, where interested folks can see machines in operation.

>Really? I'm all for letting users change the word spacing, but simply ignoring the "default" space that a type designer has (nominally) so carefully determined seems harmful.

I think "nominally" is the key here. Hrant, you realize I'm talking about *me*, not the average user (remember, I'm an elitist ;-) And I'm talking strictly about justified settings.

The main reason for my comment is this: In justified setting, the default space built into the font is largely moot. It will almost never actually occur; the space is constantly variable. I think the spacing parameters are better driven by the type size and the line length and the leading.

I prefer time-honored standards (at least as a starting point) and generally like to have optimum spacing of M/4 or M/5, minimum of M/5 or M/6, and maximum of M/3.

However, these values cannot be entered directly. H&Js in Quark and InDesign are percentages of the font's space value. So I have to determine the value of the space in order to know how to set my H&Js. For instance with Stempel Garamond, whose space value is 250, I might use 80/100/133. But for Bembo, whose space is 278, I will need to use 72/90/120 to get the same spacing.

Very inconvenient.

>Does the Quark default affect letter spacing as well? (I have InDesign)

Quark's defaults are 80/110/250 word spacing, 0/0/4 letter spacing. Notice the optimum word space is 110% of the font's space -- how idiotic. When the line cannot be acceptably hyphenated to run at optimum or tighter down to the minimum, then Quark runs out the line. After stretching the space to 250%, it will start letterspacing up to 4% wider -- even more egregious than excessive word-spacing!

(I don't fully understand the intricacies of Quark's specific algorithm, so the above is a generalization regarding when it applies which decisions between hyphenation and justification. But the results can be atrocious, regardless.)

I don't recall Indy's defaults off hand. But they are undoubtedly better. Plus, the multi-line composer can make better overall "decisions". But these are still not beyond contesting, in my limited experience with Indy.

>Kent, why do you oppose the bigger spaces? And if they are to be varied at the discression of the person setting the type, what considerations would lead to a bigger or smaller word space?

William, the case for close fitting has been made by Dowding and Bringhurst. I won't repeat them here. I remember not accepting everything by either man at face value (it's been a while since I read either), but I largely agree on this general point.

For me the bottom line is cohesion of the line and minimizing vertical drift across lines. The closer word spacing comes to the interlinear spacing, the less cohesion the line has and the more easily drift occurs. Acceptable word spacing is closely related to leading, line length, and margins.

The minimum is a function of the openness and internal fitting of the font -- a wide, open design or a loosely fit font will not tolerate as tight a minimum.

I won't go on about this. I have work to get back to. Meet me at TypeCon and ply me with sufficient food or drink and I will happily talk your ear off.

-- K.

gerald_giampa's picture

William,

You ask Kent?

Kent, why do you oppose the bigger spaces? And if they are to be varied at the discression of the person setting the type, what considerations would lead to a bigger or smaller word space?

hrant's picture

> In justified setting ... the space is constantly variable.

But it has to be based on something. And why not base it on the [nominally] optimal value for non-justified text? That said, it would be nice if apps had a way to let the user define spacing in other ways (like wrt the Em).

hhp

Nick Shinn's picture

Three related observations:

1. It's easy digitally to put in-font kerning between a space and a glyph; in combinations such as "rightdoublequote-space" and "space-T". So there are some fonts with which a sentence break can be:
...four." The...
where there is kerning between all the characters from "r" to "T"! (Myriad Condensed has a lot of this sort of thing, if I recall correctly. And I've put it into some of my fonts also.)

This a default effect determined by the font designer, not an option by the typographer. Does InDesign's "Optical Spacing" do this also?

2. Tolbert Lanston's original (1890s) justification system for the Monotype machine added incremental space between all characters, not just between words. He considered this a more sophisticated solution to the problem, however, the effect produced was not well received, so it was not put into production. Quark XPress had no qualms about making all-character justification the default setting, 90 years later.

3. When a type designer is determining the appropriate width for a particular font, it's well to consider the amount of protrusion that horizontal strokes make in characters such as "f", "r", "t", "v", "w", and "y". Sequences such as "r-space-w" and the notorious "f-space-T" make the space less apparent (eg A Tale of Two Cities, set in Bembo: compare "A-space-T" with "f-space-T").

gerald_giampa's picture

2. Tolbert Lanston's original (1890s) justification system for the Monotype machine added incremental space between all characters, not just between words. He considered this a more sophisticated solution to the problem, however, the effect produced was not well received, so it was not put into production.

Actually Nick. It was not abandoned. Very few used it. It was always an option.

There were two ways to achieve incremental additions, or subtractions to the width of type. One had nothing intended to do with word spacing.

Using a larger wedge for a more open fit, or a smaller wedge for just the opposite. Not too much smaller or mechanical problems would show. Bruce Rogers used this feature for special book projects.

Then you had unit adding, or unit subtracting. Monotypes were incredibly flexible. That does not mean all features were recommended practice for every job. Many operators never knew of their existence. Many others just never bothered with them.

That said, I never remember too many operators habitually using unit adding or subtracting except on problematical settings.

A good use for unit subtracting is kerning. A problem readily solved with a Monotype.

Unit adding would be used in instances where the operator was stuck with a sequence of unfortunate word breaks. But still, the discriminating operator would first search for a judicious use of unit subtracting often improving the setting of a line.

Unless an unfortunate line length was forced upon the operator unit adding would be a very rare thing, except when all of a sudden the author of the text decides to go on a name dropping spree. Maybe rewrite the text or cause the author to suddenly loose his friends. Don't look at me.

The wedge decision could have artistic merit, read Bruce Rogers' essay.

Tolbert Lanston was a brilliant inventor, combs, horseshoes, window sash, adding machines etc. He was not a typographer. Much to his surprise type design was a bigger issue then he realized. The same would be true of Apple computers the inventors of desktop publishing.

jim_rimmer's picture

> In justified setting . . . the spaces is constantly variable.

This is obviously true. In my experience with setting, the person setting the type must strive to come as close as possible to the most desireable space value. In good texwork this is generally thought to be a mid space (4 to the em space). The use of hyphenation makes this task easier to accomplish. Those persons who hate to see a hyphen in text must have a tough time. I would strive for even colour in spacing with the use of hyphenation, than resorting to some of the moves people use today. One that I really cannot abide is the practice of messing fit the overall fit of the type to cram a line in or to fatten one out. Now this is just a personal preference, but I would much more end up with a baggy line spacing-wise than to see some lines stand out as black bars, and others as lighter lines.

I don't know how far one can go with the adding of space between letters in a line, but for me it would have to be very little.

Goudy had a colourful comment regarding the practice of lowercase thin-spacing, but I am too much of a gent to repeat it accurately.

This thread is a good one, even if some of us have gotten a little off the rails about the space after a period.

I think there is a lot of very fine work being done today with the current computer technology. The one thing that many need to address is the subject of spacing. Let's not forget that when a person makes a new type, the thing that takes the most time is it's fitting, at least for me it is.

Jim Rimmer

gerald_giampa's picture

One that I really cannot abide is the practice of messing fit the overall fit of the type to cram a line in or to fatten one out.

Nick Shinn's picture

>Actually Nick. It was not abandoned.

I was basing my statement on an article in the Monotype Recorder, Jan-Feb 1932, which has always fascinated me. There is a comparison of the same text set two different ways, once with justification by modifying word spaces, and once by modifying letterspace.

The caption reads, "Mr Lanston's original scheme was to increase the thickness of the bodies of all the types and spaces in a complete line. He contended that the slight increase of body thickness would not be noticeable to the eye... IT EARLY BECAME OBVIOUS THAT THE ENLARGEMENT OF THE TYPE BODIES SHOULD BE CONFINED TO SPACES ONLY [my emphasis, and from this sentence I assumed that the capability had been removed]. A comparison of these [the samples] will show how Mr Lanston was mistaken in surmising that increasing body type thickness would not be noticeable, for in short measures this method of line justification is instantly obvious."

To that writer, justification by letterspacing is self-evidently wrong if noticeable -- much too shocking! -- whereas using good old word-spacing only is just fine.

gerald_giampa's picture

Nick,

I would like to read the Monotype Recorders source for this. 1932 is late in the day.

"Mr Lanston's original scheme was to increase the thickness of the bodies of all the types and spaces in a complete line. He contended that the slight increase of body thickness would not be noticeable to the eye... IT EARLY BECAME OBVIOUS THAT THE ENLARGEMENT OF THE TYPE BODIES SHOULD BE CONFINED TO SPACES ONLY

There were several reasons for this. Not always a matter of quality, but a matter of production. If a letter gets damaged, which happens, it can be replaced with another from a sorts case, or galley. Typo's were also corrected from sorts, that saved resetting lines. Fixed set widths made this very simple.

Rubber type is too gooey.

If the type characters are constant within themselves, you can not do that, the replacement will stand out like a sore thumb. Therefore the entire line should be re-cast defeating one of the fundamental strengths of the Monotype system. For that matter, the same would be true of hand set type.

And tabular matter would be impossible.

Not to mention kerning would vary, and control over outcome would be unpredictable. Remember the operator set blind.

http://lanstontype.com/MonotypeKeyboardHands.html

That said, present typography uses rubber type every day. I hate that. I agree with you, and disagree with Tolbert Lanston on that issue. That is, if indeed, he was ever so optimistic about such a bad idea.

But notice, Tolbert Lanston found reason against this idea. But applications such as Adobe PageMaker, or InDesign are rather stubbornly wrong. Some of their customers are also. I remember this discussion in the typophile forum.

Still I am not convinced that the Monotype Recorder has accurately stated this. Tolbert may have been describing a "feature" and not advocating a "radical change". Remember that proper nouns do cause a problem. For some reason unknown to me, proper nouns should not be broken. I say, "go ahead, break them."

Although, I don't.

A tool frequently used by typographical experts, was the ampersand. The ampersand was one of the typesetters best friends. Unfortunately, today's typesetters do not treat their friends well these days. The ampersand has had its' feelings hurt and has gone away.

When is the last time you attempted to use the ampersand in a text. I mean in a legitimate fashion? Try it, very disappointing. That's rubber for you.

Rubber type is not a "feature", it is a "fault". I am not saying it should never be used. I am saying it should not "have to be used". There should be choice. Those that choose bad typography should not dictate to those that do not. Unfortunately there "is more of them then there is of us." An inherent decay caused by "typographical democracy."

The tail wags the dog.

gerald_giampa's picture

Errata,

If the type characters are not constant within themselves, you can not do that, the replacement will stand out like a sore thumb.

Sorry.

William Berkson's picture

Great thread. It has raised so many questions; I've got to get a hold of Dowding and chew on all this before coming back to the issue. Meanwhile here is some more grist for the mill from Erik Spiekermann's blog:

"Text is usually set too tightly and with too big a wordspace. One of the legacies of Quark Xpress. Always set the optimum wordspace (under H&J) to 80% or less in unjustified setting. And increase tracking by at least 2 units (that

kris's picture

What a loverly coincidence! Just a few hours ago I was showing a layout of a school magazine to the client, the one I mentioned above. I told her about the double spacing no-no, and that I had fixed them to a single. She remarked that the double spacing was what she was taught in secretary school back in the day, smiled, and said that she thought I was right. What a lovely client.

kris.

hrant's picture

>> "that

Nick Shinn's picture

>Always set the optimum wordspace (under H&J) to 80% or less in unjustified setting.

This is a simplistic generalization, and bad, bad advice.

It's hard to believe that Erik knows better than all those other type designers, who (one would assume) work on the assumption that people will be using their fonts with the built-in space width, at 100%, and proportion it accordingly.

If I thought people would actually heed Erik's advice, I would pre-emptively increase the width of the spaces in my fonts. Pretty soon there would be a cold war (a space race), with the spaces in fonts getting bigger and bigger, and Erik's recommended optimum getting smaller and smaller.

hrant's picture

No, because between Erik and Quark, we need to just hold our ground. ;->

hhp

gerald_giampa's picture

I would rather ask than search.

Your space width, what method do you use to determine their size?

Nick Shinn's picture

>Your space width, what method do you use to determine their size?

First, give it a value that looks about right in the metrics window.
Than, after a font has been generated, test it in a Quark document. If the space doesn't look right there, I can vary its horizontal scaling in Quark H&Js (according to the method Erik described), comparing paragraphs that are the same except for the space width; then make the corresponding adjustment in the font.

theorosendorf's picture

Here's a discussion about double/single spacing on Blogdorf:
http://rosendorf.us/blogdorf/archive/2005/07/14/449.aspx

~
Theodore Rosendorf
http://rosendorf.us

Stephen Coles's picture

Thank you for this topic, Daniel.

I have long been bothered by the double space phenomenon.
We were taught to do it in school, and even then, without any
knowledge or interest in typography, it seemed senseless to me.
No book or magazine or newspaper I could find had two spaces
after each sentence. One space clearly set apart professionally
typeset material from that made by uninformed humans. I vowed
to never type those two spaces again. Satisfied in my little
rebellion against the educational system, I was later pleased to
find that professional typography confirmed I was justified.
That epiphany was part of what got me so interested in
typography in the first place.

But that's a bit of a digression from my main point which is that
the result of this misteaching is a lot of correspondence created
on a computer with proportional fonts is still rife with two spaces.
Few people seem to know any better. Robin Williams tried to
educate the uninformed masses with her "The Mac Is Not
a Typewriter" a few years back ("The PC Is Not A Typewriter"
came out soon after), but even a book that was marketed to an
audience beyond graphic designers reached too few.

There is some hope. It's interesting to note that HTML will not
accept two spaces in a row and will automatically remove the
second. Nothing on the web is double-spaced - even the stuff
that isn't part of a publication (like those I mentioned). Maybe
now, with the advent of online forums and an increasingly web-
aware population, people will start to get the idea that the
practice isn't necessary. I sure hope so. Unfortunately, most
folks probably aren't observant enough to notice that their extra
spaces are being zapped.

A question that maybe some Typophiles with kids can answer:
do they still teach this in schools?

(This post was written with two spaces after each sentence as I
was taught in all 12 years of my primary and secondary education.
Half those spaces will be generously removed by the great
potential re-educator: HTML.)

Stephen Coles's picture

"Call me a heretic, but I prefer two spaces after periods. I read
a lot, and some books seem to have NO space at all after periods"


So why is one space not just right? Like it's been mentioned, the
space above the period plus one space is enough to set
sentences apart. If you believe that more space is needed, you
and I have very different notions of even color and readability.

Syndicate content Syndicate content