oversized margins
Wolfgang Jeltsch
I use LaTeX' article class with the a4paper option. I get really large
margins at all page borders, especially to the left and to the right. I
suppose, the latter is because LaTeX reserves room vor margin paragraphs
made with \marginpar. But I don't use this feature.
Is there a standard way of reducing the size of the margins to more
appropriate values? Of course, I could change them all by hand but I'm
looking for a solution (perhaps a special package) which is made by people
who know more about typesetting than me.
Thanks in advance.
Wolfgang
Ulrike Fischer
http://www.dante.de/faq/de-tex-faq/html/de-tex-faq.html
5.4.1 Gibt es neben den Standard-LaTeX-Klassen auch Klassen mit einem
,europäischeren` Design?
Use Koma-script or memoir.
--
Ulrike Fischer
e-mail: zusätzlich meinen Vornamen vor dem @ einfügen.
e-mail: add my first name between the news and the @.
Tristan Miller
In article <boqtbo$1h8l4f$1...@ID-77306.news.uni-berlin.de>, Wolfgang
Jeltsch wrote:
> I use LaTeX' article class with the a4paper option. I get really large
> margins at all page borders, especially to the left and to the right.
> I suppose, the latter is because LaTeX reserves room vor margin
> paragraphs made with \marginpar. But I don't use this feature.
This is not unique to A4 paper. I always supposed the large margins
were because people find it difficult and annoying to read 10-point (or
even 12-point) text all the way across a page. Take a look at
practically any profesionally-typeset book, magazine, or newspaper you
own and you'll see that the columns are rarely wider than what you get
with LaTeX's defaults. People have been conditioned by the default
settings on popular word processors to think that 2.5-centimetre
margins on A4 or US letter paper are "normal" and that anything greater
is a waste of paper.
> Is there a standard way of reducing the size of the margins to more
> appropriate values?
When I'm required to use narrower margins, I use the "fullpage" package,
which seems to automatically take care of things. However, I consider
typesetting a single column of text with this package to make for
uncomfortable reading.
Regards,
Tristan
--
_
_V.-o Tristan Miller [en,(fr,de,ia)] >< Space is limited
/ |`-' -=-=-=-=-=-=-=-=-=-=-=-=-=-=-= <> In a haiku, so it's hard
(7_\\ http://www.nothingisreal.com/ >< To finish what you
Юрий Зуев/Yuri Zuev
not a serious problem to read wide pages.
I use \oddsidemargin, \evensidemargin, \topmargin commands in the
preamble. The other solution above (with fullpage package) may be better
though.
Yuri
Aditya Mahajan
"Wolfgang Jeltsch" <jel...@tu-cottbus.de> wrote in message
news:boqtbo$1h8l4f$1...@ID-77306.news.uni-berlin.de...
Have a look at geometry.sty It is the easiest way of changing margins.
--
Aditya Mahajan
Rowland McDonnell
The thing about these margins that you see is that they *are* an
appropriate size. The problem is the paper size, not LaTeX's
typesetting.
It's like this: when it comes to printed blocks of text, there's range
of line lengths which allow the text to be read easily by most people.
More or less than that range, and it's harder to read. What the upper
limit actually *is* is a matter of debate, but if you stick to not more
than about 75 characters of normal prose, you'll be okay[1]. Check
LaTeX's standard classes: they actually slightly exceed this limit. The
problem is that A4 paper is just too wide for single columns of text in
10pt-12pt sizes. It's great for old-fashioned typewriting (take a look
at the sheer width of typescript), but rotten for proper printing founts
- because you get margins that are just too wide, as you've noticed.
The problem is that you shouldn't reduce the size of the margins,
because that'll make the text awkward to read even though the text looks
better on the page from a distance.
The only sane way round this is to use two columns, I reckon[1].
Mind you, I really don't like the way the standard LaTeX classes
apportion space around the text block - the suggestions to use
alternatives to the standard classes and bolt-on packages are good ones.
But don't overdo the size of \textwidth, eh?
Rowland.
[1] A common way of specifying the range that I've read in typography
books is to say `between 1.5 and 2.5 alphabets'. That is, you should
measure the width of the alphabet typeset in the fount you're using, and
make the lines between 1.5 and 2.5 as wide as that. Hmm. Strange that
I never thought to check whether they meant upper or lower case or both.
Rowland.
[1] Insane people might just revert to foolscap folio; I still have
some, and my slightly dodgy rmpage package knows about the size by name.
--
Remove the animal for email address: rowland....@dog.physics.org
PGP pub key 0x62DCCA78 Sorry - the spam got to me
http://www.mag-uk.org
UK biker? Join MAG and help keep bureaucracy at bay
Brian Blackmore
> Wolfgang Jeltsch <jel...@tu-cottbus.de> wrote:
>> I use LaTeX' article class with the a4paper option. I get really large
>> margins at all page borders, especially to the left and to the right. I
>> suppose, the latter is because LaTeX reserves room vor margin paragraphs
>> made with \marginpar. But I don't use this feature.
>>
>> Is there a standard way of reducing the size of the margins to more
>> appropriate values? Of course, I could change them all by hand but I'm
>> looking for a solution (perhaps a special package) which is made by people
>> who know more about typesetting than me.
> The thing about these margins that you see is that they *are* an
> appropriate size. The problem is the paper size, not LaTeX's
> typesetting.
Thank you Rowland for telling the users that the style designers knew what
they were doing and things shouldn't be changed. :) It gets old after a
while to think that all these people asking these questions over and over
and over again really people with doctorates in graphics design or
background in typography. Alas, the drag-and-clickity interface of
you-know-what has left people with this idea that they can do whatever
they want. I guess they can; I guess that's why nothing looks good.
> It's like this: when it comes to printed blocks of text, there's range
> of line lengths which allow the text to be read easily by most people.
> More or less than that range, and it's harder to read. What the upper
> limit actually *is* is a matter of debate, but if you stick to not more
> than about 75 characters of normal prose, you'll be okay[1]. Check
> LaTeX's standard classes: they actually slightly exceed this limit.
There is something else important to happy-TeX though, and that's the
ability to construct non-bad lines. If the lines are way too narrow, TeX
is going to burn your CPU trying to find happy line breaks and
hyphenations, etc., and in the end it's just going to break down and start
giving over/underfull box warnings. Sticking with something slightly
larger (than 65 characters wide) is probably giving the engine a bit more
breathing room to construct line breaks.
To drive home the point about comfortable widths, I will mention that this
is one of the primary reasons that thinking PDF is an acceptable online
browsing format is just wrong wrong wrong, and remark that my comfortable
reading width on a screen changes throughout the day. On the printed
page, I can adhere to a width of about 70 characters, and I drop to two
columns when taking notes by hand on US Letter sized paper, but on the
screen I usually read and write at around 150 characters per line.
Now everyone shouts that I'm absurd. I will just note one study that
found that screen reading at 100 characters per line was a bit faster than
everyone's happy 75[1]. By extension, it is reasonable to read at wider
widths than that. :)
> The problem is that A4 paper is just too wide for single columns of text
> in 10pt-12pt sizes. ...
I definitely agree with you here, but will note that if a large chunk of
the text is going to be displayed-style mathematics or mathematics that is
naturally more wide, then one could get away with stretching the text
width somewhat.
> Mind you, I really don't like the way the standard LaTeX classes
> apportion space around the text block - the suggestions to use
> alternatives to the standard classes and bolt-on packages are good ones.
> But don't overdo the size of \textwidth, eh?
It always immediately strikes me from a distance though, that a page has
been printed with LaTeX because it immediately looks ugly :-D There's
something about that text block that just shrieks nasty things in my
direction.
For my homework style, I had to sit down for several hours to come up with
a good text block (i.e., a text block that I like and think is reasonable,
though that may not be entirely so). That included considerations of
utility and readability and useability, and those are often at odds with
one another.
> [1] A common way of specifying the range that I've read in typography
> books is to say `between 1.5 and 2.5 alphabets'. That is, you should
> measure the width of the alphabet typeset in the fount you're using, and
> make the lines between 1.5 and 2.5 as wide as that. Hmm. Strange that
> I never thought to check whether they meant upper or lower case or both.
I am not a typographer, but I believe that refers to lowercase. The table
in Bringhurst [2] uses lowercase alphabet length, and seems to prefer
columns with 60 to 69 characters.
[1] "Effects of Line Lengths and Method of Movement on Patterns of Reading
From Screen". Visible Language 32.2. Page 166.
http://www.id.iit.edu/visiblelanguage/
[2] Bringhurst, Robert. The Elements of Typographic Style. Version 2.4.
Hartley & Marks, Publishers. Vancouver. 2001. p. 29.
--
Brian Blackmore
blb8 at po dot cwru dot edu
Giuseppe Bilotta
> Now everyone shouts that I'm absurd. I will just note one study that
> found that screen reading at 100 characters per line was a bit faster than
> everyone's happy 75[1]. By extension, it is reasonable to read at wider
> widths than that. :)
>
> From Screen". Visible Language 32.2. Page 166.
> http://www.id.iit.edu/visiblelanguage/
I have not yet read the article, since I'm offline at the moment, but
does it take into consideration the fact that most commonly used
screens are in 4:3 ratio and thus using only 75 characters per line
at a decent size needs more scrolling? Just by turning my monitor 90
degrees (I'm one of the lucky guys who can do it) I noticeably
improve my screen-reading performance.
--
Giuseppe "Oblomov" Bilotta
Can't you see
It all makes perfect sense
Expressed in dollar and cents
Pounds shillings and pence
(Roger Waters)
Rowland McDonnell
> Rowland McDonnell <real-addr...@flur.bltigibbet> wrote:
> > Wolfgang Jeltsch <jel...@tu-cottbus.de> wrote:
> >> I use LaTeX' article class with the a4paper option. I get really large
> >> margins at all page borders, especially to the left and to the right. I
> >> suppose, the latter is because LaTeX reserves room vor margin paragraphs
> >> made with \marginpar. But I don't use this feature.
> >>
> >> Is there a standard way of reducing the size of the margins to more
> >> appropriate values? Of course, I could change them all by hand but I'm
> >> looking for a solution (perhaps a special package) which is made by people
> >> who know more about typesetting than me.
>
> > The thing about these margins that you see is that they *are* an
> > appropriate size. The problem is the paper size, not LaTeX's
> > typesetting.
>
> Thank you Rowland for telling the users that the style designers knew what
> they were doing and things shouldn't be changed. :)
<chuckle> I wouldn't quite go *that* far - those classes aren't great,
but they were actually properly designed, so you shouldn't change
anything unless you understand *why* the LaTeX class was the way it was,
and how to ensure that your changes are in fact an overall improvement.
Typography is frankly a right bugger - you've got to balance the
requirements of good looks at a distance, good looks close up, and ease
of use - typographers need to take all three into account
simultaneously. Non experts like me who have read a bit on the subject
need to think very hard in order to do that; and unless you've done some
study of the field, you don't even know what it is you need to pay
attention to, so you're pretty much guaranteed to screw up.
(one problem is that I'm pretty sure that most people *can't* run three
aesthetic senses simultaneously. I suspect that good typographers can
switch between the three views very rapidly. I find that I have to take
an overnight break to get sufficient distance from my previous outlook -
and that doesn't half slow things down, especially when you're just
trying to fiddle around and get a feel for `how things work')
Having said that, you might find it useful to read the documentation for
the memoir class - it's not a full guide to typography or anything, but
it's got a lot of very interesting and useful things to say about how to
arrange text on your pages. This is required in the docs, because the
memoir class is designed as an easy-to-modify general purpose LaTeX
class meant to stand in for article, report, and book - more or less.
> It gets old after a
> while to think that all these people asking these questions over and over
> and over again really people with doctorates in graphics design or
> background in typography.
I wouldn't trust anyone with a graphics design background myself.
> Alas, the drag-and-clickity interface of
> you-know-what has left people with this idea that they can do whatever
> they want. I guess they can; I guess that's why nothing looks good.
The problem is more that they are *ignorant*. I used WPs and then LaTeX
for *years* before starting to learn about typography - and that was
only because I took the trouble to visit the library and read some
books. What that meant was that before I did so, pretty much all my WP
output and quite a lot of my LaTeX output was very rough.
But what can one do? I've taught typing and WP use professionally to
official exam standards - and those exams, which are the standard exams
you take in the UK if you want to be an office drone all your life -
those exams do not mention *anything* about how to set up your WP to
produce easy-to-read output. There's lot about making sure you've got 1
inch margins in some of the standard textbooks - as if we were still
using typewriters on foolscap, fer gawd's sake.
There's no hope until this sort of thing changes: WP publishers should
be producing style guides, and brief introductions to typography (and
yes, you *can* write something useful on the subject that's quick and
easy to read while being very, very useful). And the people who design
education and training for those in need of typing instruction ought to
add a touch on basic typgraphy - keep the lines and pages short enough
but not too short (with examples), use sizes like *this*, avoid sanserif
unless the boss insists, and don't use lots of founts.
> > It's like this: when it comes to printed blocks of text, there's range
> > of line lengths which allow the text to be read easily by most people.
> > More or less than that range, and it's harder to read. What the upper
> > limit actually *is* is a matter of debate, but if you stick to not more
> > than about 75 characters of normal prose, you'll be okay[1]. Check
> > LaTeX's standard classes: they actually slightly exceed this limit.
>
> There is something else important to happy-TeX though, and that's the
> ability to construct non-bad lines. If the lines are way too narrow, TeX
> is going to burn your CPU trying to find happy line breaks and
> hyphenations, etc., and in the end it's just going to break down and start
> giving over/underfull box warnings. Sticking with something slightly
> larger (than 65 characters wide) is probably giving the engine a bit more
> breathing room to construct line breaks.
Actually, TeX's pretty good at breaking lines - and if you *do* want
short lines, you have to tell TeX to do what all typesetters do when
faced with (for example) newspaper columns: accept bloody awful
typesetting as the price you have to pay from time to time. In the case
of TeX, this means you ought to change the paragraph building parameters
so that it will accept worse solutions.
> To drive home the point about comfortable widths, I will mention that this
> is one of the primary reasons that thinking PDF is an acceptable online
> browsing format is just wrong wrong wrong, and remark that my comfortable
> reading width on a screen changes throughout the day. On the printed
> page, I can adhere to a width of about 70 characters, and I drop to two
> columns when taking notes by hand on US Letter sized paper, but on the
> screen I usually read and write at around 150 characters per line.
>
> Now everyone shouts that I'm absurd.
I don't.
> I will just note one study that
> found that screen reading at 100 characters per line was a bit faster than
> everyone's happy 75[1]. By extension, it is reasonable to read at wider
> widths than that. :)
Screen reading is significantly different to paper reading. I suspect
that it's harder to whizz back to the start of a line on a screen than
on paper - which might mean that excessively long lines don't introduce
a slow-down because you're suffering one anyway.
> > The problem is that A4 paper is just too wide for single columns of text
> > in 10pt-12pt sizes. ...
>
> I definitely agree with you here, but will note that if a large chunk of
> the text is going to be displayed-style mathematics or mathematics that is
> naturally more wide, then one could get away with stretching the text
> width somewhat.
I reckon it's a better idea to have two column text in those
circumstances - and if you really do have wide maths, run it across two
columns. I expect plenty of people disagree with me.
> > Mind you, I really don't like the way the standard LaTeX classes
> > apportion space around the text block - the suggestions to use
> > alternatives to the standard classes and bolt-on packages are good ones.
> > But don't overdo the size of \textwidth, eh?
>
> It always immediately strikes me from a distance though, that a page has
> been printed with LaTeX because it immediately looks ugly :-D There's
> something about that text block that just shrieks nasty things in my
> direction.
One problem with the standard LaTeX classes is that they assume you will
have a header and a footer. If you don't have a header, the text block
is positioned too far down the page. And the text block is also far too
narrow if you're using 10pt or 11pt with cmr and the standard classes
(it's even worse if you're using Times at those sizes and you've reduced
\textwidth appropriately to take into account the narrower set of
Times).
Another problem is the use of Computer Modern Bold Extended for
sectional headers, and the `feel' is very - well, 1980s USA, which was
fine for the 1980s Americans who came up with 'em but grates on me at
least.
You might find that the KOMA-script classes are more to your taste.
> For my homework style, I had to sit down for several hours to come up with
> a good text block (i.e., a text block that I like and think is reasonable,
> though that may not be entirely so). That included considerations of
> utility and readability and useability, and those are often at odds with
> one another.
You might find the memoir class very useful. And while I hesitate to
mention it (because I've still not got round to sorting out the
documentation and I really need to get on with a complete re-write), my
rmpage packge on CTAN is something you might find handy for playing
around with the text block size.
(the released version of rmpage is a *lot* more clever than the early
versions which have long since died and been cremated; but it's a
lumbering, eccentric dinosaur with crap docs that I really ought to
rationalise and, erm, change so that it tends to produce excellent
results rather than what I happened to think was about right at the
time)
> > [1] A common way of specifying the range that I've read in typography
> > books is to say `between 1.5 and 2.5 alphabets'. That is, you should
> > measure the width of the alphabet typeset in the fount you're using, and
> > make the lines between 1.5 and 2.5 as wide as that. Hmm. Strange that
> > I never thought to check whether they meant upper or lower case or both.
>
> I am not a typographer, but I believe that refers to lowercase.
Me too - but it's strange that I never thought to check in the books I
was reading.
> The table
> in Bringhurst [2] uses lowercase alphabet length, and seems to prefer
> columns with 60 to 69 characters.
>
> [1] "Effects of Line Lengths and Method of Movement on Patterns of Reading
> From Screen". Visible Language 32.2. Page 166.
> http://www.id.iit.edu/visiblelanguage/
>
> [2] Bringhurst, Robert. The Elements of Typographic Style. Version 2.4.
> Hartley & Marks, Publishers. Vancouver. 2001. p. 29.
Thank you for this.
Rowland.
Brian Blackmore
> Brian Blackmore wrote:
>> Now everyone shouts that I'm absurd. I will just note one study that
>> found that screen reading at 100 characters per line was a bit faster than
>> everyone's happy 75[1]. By extension, it is reasonable to read at wider
>> widths than that. :)
>>
>> [1] "Effects of Line Lengths and Method of Movement on Patterns of Reading
>> From Screen". Visible Language 32.2. Page 166.
>> http://www.id.iit.edu/visiblelanguage/
> I have not yet read the article, since I'm offline at the moment, but
> does it take into consideration the fact that most commonly used
> screens are in 4:3 ratio and thus using only 75 characters per line
> at a decent size needs more scrolling? Just by turning my monitor 90
> degrees (I'm one of the lucky guys who can do it) I noticeably
> improve my screen-reading performance.
I do not recollect the complete text layout, but there was certainly no
side-to-side scrolling in the experiment. The movement was restricted to
either full-pagination or single line scrolling and, if memory serves,
both setups allowed backward motion. The time and memory retention was
recorded.
I suspect that I would have done quite poorly in their tests because I am
rather particular about having a black background and I tend toward
cursor-assisted tracking and line-by-line scrolling, though I fully
paginate in certain cases.
The reason I recorded a reference of sorts to the study --- sorry I don't
recall the full published information, but it might have been that the
printed journal only gave the URL (?) --- was that they showed that a
sampling of `normal people' (whatever that is) could actually read for
retention at a reasonable speed with large column widths.
This is one of those pet-peeves of mine because it effects everyone all
the time. This is why, IMHO, PDF is NOT A BROWSER FORMAT because I can't
resize on the fly. This is why HTML should NOT come with stupid <TABLE
width=550> tags surrounding it. I think it would be interesting to have
one of the web-statistics sites collate information about how many sites
do this, use Javascript, flash, etc., just to see how much beautiful HTML
still exists. I'm hoping that the silly people can learn and use CSS in
the near future so I can happily browse the pages at my chosen,
comfortable width, perhaps on a wireless, handheld device, perhaps on a
fourteen-foot wall screen, etc.
The one place I still `adhere' is on newsgroups --- I'll still stick with
widths under 80; I don't know why, but I do. But I'm on a TeX newsgroup
so, happily, a blank line equals a paragraph boundary. :)
Rowland McDonnell
[snip]
> The reason I recorded a reference of sorts to the study --- sorry I don't
> recall the full published information, but it might have been that the
> printed journal only gave the URL (?) --- was that they showed that a
> sampling of `normal people' (whatever that is) could actually read for
> retention at a reasonable speed with large column widths.
I don't find retention the problem, but I usually re-size to a small
column width when Web browsing because I find that wide columns make it
hard for me to find the start of the next line. Then again, I often use
really huge widths for text editing LaTeX code especially - that's not a
problem because of the extreme raggedness of the right hand edge.
[snip]
> The one place I still `adhere' is on newsgroups --- I'll still stick with
> widths under 80; I don't know why, but I do.
[snip]
One not very good reason is that I'll whinge if you don't, because of a
particularly fascist newsreader which won't let me post with over-long
lines - and that means over-long lines in quoted text too.
Rowland.
Lucian Wischik
>This is one of those pet-peeves of mine because it effects everyone all
>the time. This is why, IMHO, PDF is NOT A BROWSER FORMAT because I can't
>resize on the fly.
Yes you can...
A "tagged" PDF contains structural markup like HTML does. That enables
"reflowing" when the PDF-viewer window size changes. Adobe are keen on
this for PDA use. Tagged PDF is generated automatically by Word2000/XP
and in PageMaker7. I don't know what other tools generate it.
http://www.adobe.com/products/acrobat/pdfs/CreateAccessibleAdvanced.pdf
--
Lucian
Danie Els
> I agree it is an utter waste of paper, just like double-spacing. It's
> not a serious problem to read wide pages.
>
It is not all about stuffing as much info on a page as possible. Wide
margins are not there only for margin pars, but it is also a place to
hold the book (for your thumbs). One of the main reasons for the wide
margins is that the book looks good as a whole. To quote Bringhurst: "
Perhaps fifty per cent of the character and integrity of a printed
page lies in its letterforms. Much of the other fifty per cent resides
in its margins."
The one thing I don't like about the LaTeX page layout is the fact
that the textwidth is fix to plus-minus 75 character of Computer
Modern font for all font sizes. For my own purposes (A4 pages) I
normly fix the text width and height and then select the font size to
give a decent number of character.
On a A4 page, if you select a type block with width 2/3 page width and
the height to width ration equal to the golden ratio, you find quite a
pleasing layout (better than the latex one). A foredge and top marging
of 1/9 page width gives nice wide margins but running head are
cramped. I prefer widths of 1/8. If you use this fixed layout you will
find that you will need fontsizes of Palatio 11 pt, CM 12 pt, Times
12 or 14 pt :-o , Lucida 11 pt, etc.
So, my friend, stop forcing to much info on one page and look at the
artistic (very conservative) value of the total layout.
Danie Els
(dnjels at sun dot ac dot za)
Rowland McDonnell
> Yuri Zuev <z...@ohio-state.edu> wrote:
> > I agree it is an utter waste of paper, just like double-spacing. It's
> > not a serious problem to read wide pages.
No, but it does slow you down a bit and it does make the `reading
experience' (yech) less pleasant.
> It is not all about stuffing as much info on a page as possible. Wide
> margins are not there only for margin pars, but it is also a place to
> hold the book (for your thumbs). One of the main reasons for the wide
> margins is that the book looks good as a whole. To quote Bringhurst: "
> Perhaps fifty per cent of the character and integrity of a printed
> page lies in its letterforms. Much of the other fifty per cent resides
> in its margins."
There's a lot of rubbish talked about typography.
> The one thing I don't like about the LaTeX page layout is the fact
> that the textwidth is fix to plus-minus 75 character of Computer
> Modern font for all font sizes.
That's not the LaTeX page layout; it's just what you get with the
standard LaTeX classes. My rmpage package (available on CTAN but with
unfinished docs and mad design 'cos I'm mad myself; I use it for
everything except perhaps the washing up) made an attempt to deal with
this particular problem - it'll measure the fount in use if you ask it
to, and will set \textwidth based on that, with permutations available
to make the width smaller or larger than the `usual easy-to-read width'.
> For my own purposes (A4 pages) I
> normly fix the text width and height and then select the font size to
> give a decent number of character.
<curious> Why not do your fount selection as an integral part of your
document design? That is, say `I want a wide measure in one column, so
I'd better pick a wide setting typeface', sort of thang.
> On a A4 page, if you select a type block with width 2/3 page width and
> the height to width ration equal to the golden ratio, you find quite a
> pleasing layout (better than the latex one).
Some argue that decoupling the text body aspect ratio from the
surrounding paper aspect ratio, that you get a non-optimal result.
Others, obviously, disagree. Me? I have no fixed opinion, but also
often type `ration' instead of `ratio'.
> A foredge and top marging
> of 1/9 page width gives nice wide margins but running head are
> cramped. I prefer widths of 1/8. If you use this fixed layout you will
> find that you will need fontsizes of Palatio 11 pt, CM 12 pt, Times
> 12 or 14 pt :-o , Lucida 11 pt, etc.
Computer Modern Roman is about the same width as Palatino; Times is much
narrower. Using rmpage to do the measuring of the average width of a
character in the fount concerned, I get (using 2/3 x 210mm =
398.3386...pt):
ppl 12pt is 5.26640 pt/char, giving 75.6 chars across 2/3 of A4
cmr 12pt is 5.16339 pt/char, giving 77.1 chars across 2/3 of A4
ptm 12pt is 4.76569 pt/char, giving 83.6 chars across 2/3 of A4
I wouldn't use Times at all across that measure - 12pt gives far too
many chars across the width, and 14pt is too big for body copy unless
you're aiming at a special audience. 12pt is about right for Palatino
at that \textwidth and I expect 12pt for cmr is fine too at that width.
> So, my friend, stop forcing to much info on one page and look at the
> artistic (very conservative) value of the total layout.
Good idea - but read some typography books before changing the layout.
You probably won't learn `how to do it' (I didn't), but you will learn
`what to look for' (I did).
btw, rather than saying `artistic', I'd say `aesthetic'; but that's
because I'm with Oscar Wilde in thinking that all art is by definition
useless, and we don't want useless page layouts, do we?
Brian Blackmore
> Brian Blackmore <bl...@po.cwru.edu> wrote:
>> Rowland McDonnell <real-addr...@flur.bltigibbet> wrote:
>> > Wolfgang Jeltsch <jel...@tu-cottbus.de> wrote:
>> >> I use LaTeX' article class with the a4paper option. I get really large
>> >> margins at all page borders, especially to the left and to the right.
>> >> I suppose, the latter is because LaTeX reserves room vor margin
>> >> paragraphs made with \marginpar. But I don't use this feature.
>> >>
>> >> Is there a standard way of reducing the size of the margins to more
>> >> appropriate values? Of course, I could change them all by hand but I'm
>> >> looking for a solution (perhaps a special package) which is made by
>> >> people who know more about typesetting than me.
>>
>> > The thing about these margins that you see is that they *are* an
>> > appropriate size. The problem is the paper size, not LaTeX's
>> > typesetting.
>>
>> Thank you Rowland for telling the users that the style designers knew what
>> they were doing and things shouldn't be changed. :)
> ... they were actually properly designed ... Non experts like me who
> have read a bit on the subject need to think very hard in order to do
> that; and unless you've done some study of the field, you don't even
> know what it is you need to pay attention to, so you're pretty much
> guaranteed to screw up.
My point was that
they were designed
properly and that
the bulk of users
asking to change
margins haven't even
"read a bit on the
subject", aren't
thinking about what
they're doing, don't
know what they need
to pay attention to,
and are pretty much
screwing up.
Instead of allowing
these types of bad
decisions, I find
myself often asking
the people what they
want to do and why,
and try to get them
to understand that
it's bad. :)
> (one problem is that I'm pretty sure that most people *can't* run three
> aesthetic senses simultaneously. I suspect that good typographers can
> switch between the three views very rapidly. I find that I have to take
> an overnight break to get sufficient distance from my previous outlook
> ...
This is why certain styles have taken me several hours to create, of
course, and several hours to correct and/or improve, and hours to monitor
and alter, and hours and hours and hours. My overnight break tends to
occur as a matter of fact, simply because I'd pass out before projects
were finished if I were trying to get them all done at once.
> Having said that, you might find it useful to read the documentation for
> the memoir class - it's not a full guide to typography or anything, but
> it's got a lot of very interesting and useful things to say about how to
> arrange text on your pages. ...
Perhaps I will at some point. It might help me gain some insight into the
minds of the LaTeX crowd.
>> It gets old after a while to think that all these people asking these
>> questions over and over and over again really people with doctorates
>> in graphics design or background in typography.
> I wouldn't trust anyone with a graphics design background myself.
They have `authority', you know, those people with degrees in graphics
design, etc.; at the least, they think they do. But I suspect that most
users I see asking questions have gotten their authority from Word: "It
does it that way, so it must be right!"
>> Alas, the drag-and-clickity interface of you-know-what has left people
>> with this idea that they can do whatever they want. I guess they can;
>> I guess that's why nothing looks good.
> The problem is more that they are *ignorant*. I used WPs and then LaTeX
> for *years* before starting to learn about typography ...
This is true; a good number of people have a lot that they just don't care
about.
> But what can one do? I've taught typing and WP use professionally to
> official exam standards ... There's lot about making sure you've got 1
> inch margins in some of the standard textbooks - as if we were still
> using typewriters on foolscap, fer gawd's sake.
Yeah, and here at a university I've received a rather cold reception on
the TeX/LaTeX front. It is my understanding that universities are
supposed to be great havens for happy software such as mentioned, and for
collections of *nix operating systems. I guess the problem at this
particular university is that there are people and buildings named `Gates'
and Redmond is 40 miles away.
> There's no hope until this sort of thing changes: WP publishers should
> be producing style guides, and brief introductions to typography ...
> And the people who design education and training for those in need of
> typing instruction ought to add a touch on basic typgraphy ...
Blech, WP styles. Well, I would suspect that they didn't do it before
because it would have required reading a 2000 page manual. Then again,
Computers and Typography is 2000 pages, so I'm not sure what that means.
I think the majority of people who do the training probably don't know any
better either, so it's likely a self-propogating problem.
>> > It's like this: when it comes to printed blocks of text, there's range
>> > of line lengths which allow the text to be read easily by most people.
>> > More or less than that range, and it's harder to read. What the upper
>> > limit actually *is* is a matter of debate, but if you stick to not more
>> > than about 75 characters of normal prose, you'll be okay[1]. Check
>> > LaTeX's standard classes: they actually slightly exceed this limit.
>>
>> There is something else important to happy-TeX though, and that's the
>> ability to construct non-bad lines. If the lines are way too narrow, TeX
>> is going to burn your CPU trying to find happy line breaks and
>> hyphenations, etc., and in the end it's just going to break down and start
>> giving over/underfull box warnings. Sticking with something slightly
>> larger (than 65 characters wide) is probably giving the engine a bit more
>> breathing room to construct line breaks.
> Actually, TeX's pretty good at breaking lines ...
Agreed.
>> To drive home the point about comfortable widths, I will mention that this
>> is one of the primary reasons that thinking PDF is an acceptable online
>> browsing format is just wrong wrong wrong, and remark that my comfortable
>> reading width on a screen changes throughout the day. On the printed
>> page, I can adhere to a width of about 70 characters, and I drop to two
>> columns when taking notes by hand on US Letter sized paper, but on the
>> screen I usually read and write at around 150 characters per line.
>>
>> Now everyone shouts that I'm absurd.
> I don't.
Shouted something else instead?
>> I will just note one study that found that screen reading at 100
>> characters per line was a bit faster than everyone's happy 75[1]. By
>> extension, it is reasonable to read at wider widths than that. :)
> Screen reading is significantly different to paper reading. I suspect
> that it's harder to whizz back to the start of a line on a screen than
> on paper - which might mean that excessively long lines don't introduce
> a slow-down because you're suffering one anyway.
I think when one has a cursor that it is quite easy to track the
beginnings of the lines, hence this whizz-back is not a problem. What the
report showed was that reading at 100 characters was faster (statistically
so, if memory serves). This was amusing to me, since there are certain
instances where I feel I'm reading much faster at 150 characters wide.
>> > The problem is that A4 paper is just too wide for single columns of text
>> > in 10pt-12pt sizes. ...
>>
>> I definitely agree with you here, but will note that if a large chunk of
>> the text is going to be displayed-style mathematics or mathematics that is
>> naturally more wide, then one could get away with stretching the text
>> width somewhat.
> I reckon it's a better idea to have two column text in those
> circumstances - and if you really do have wide maths, run it across two
> columns. I expect plenty of people disagree with me.
<shurg>
>> > Mind you, I really don't like the way the standard LaTeX classes
>> > apportion space around the text block - the suggestions to use
>> > alternatives to the standard classes and bolt-on packages are good ones.
>> > But don't overdo the size of \textwidth, eh?
>>
>> It always immediately strikes me from a distance though, that a page has
>> been printed with LaTeX because it immediately looks ugly :-D There's
>> something about that text block that just shrieks nasty things in my
>> direction.
> One problem with the standard LaTeX classes is that they assume you will
> have a header and a footer. ...
> Another problem is the use of Computer Modern Bold Extended for
> sectional headers, and the `feel' is very - well, 1980s USA, which was
> fine for the 1980s Americans who came up with 'em but grates on me at
> least.
Perhaps this is what is really happening to me. I wasn't doing anything
typographic in the 80s, so I wouldn't have noticed. :\
>> For my homework style, I had to sit down for several hours to come up with
>> a good text block (i.e., a text block that I like and think is reasonable,
>> though that may not be entirely so). That included considerations of
>> utility and readability and useability, and those are often at odds with
>> one another.
> You might find the memoir class very useful. And while I hesitate to
> mention it (because I've still not got round to sorting out the
> documentation and I really need to get on with a complete re-write), my
> rmpage packge on CTAN is something you might find handy for playing
> around with the text block size.
To be quite frank about it, I find that TeX is very useful, but maybe I
should still read some of the documentation and design decisions that were
made for memoir and KOMA.
I seem to recall noting rmpage at one point, when I was searching for
something, so it has likely helped a good collection of people. I don't
recall exactly what it does, but I think it was mentioned at another spot
in this thread, and I heard a mention of lowercase widths and
characters-per-line computations. I've been known to have TeX tell me
this type of stuff so, if this is indeed the type of thing it does, you
and I are of like mind (we certainly seem to be alike in our propensity
for verbosity).
>> > [1] A common way of specifying the range that I've read in typography
>> > books is to say `between 1.5 and 2.5 alphabets'. That is, you should
>> > measure the width of the alphabet typeset in the fount you're using, and
>> > make the lines between 1.5 and 2.5 as wide as that. Hmm. Strange that
>> > I never thought to check whether they meant upper or lower case or both.
>>
>> I am not a typographer, but I believe that refers to lowercase.
> Me too - but it's strange that I never thought to check in the books I
> was reading.
Details. Hrmph.
>> The table in Bringhurst [2] uses lowercase alphabet length, and seems
>> to prefer columns with 60 to 69 characters.
>>
>> [1] "Effects of Line Lengths and Method of Movement on Patterns of Reading
>> From Screen". Visible Language 32.2. Page 166.
>> http://www.id.iit.edu/visiblelanguage/
>>
>> [2] Bringhurst, Robert. The Elements of Typographic Style. Version 2.4.
>> Hartley & Marks, Publishers. Vancouver. 2001. p. 29.
> Thank you for this.
Alas, there are still people who don't know Bringhurst, so it is necessary
to include the reference.
Maybe it's not as bad as I think, but perhaps it's time I stop writing to
newsgroups when I'm in a bad mood.
Donald Arseneau
> My point was that
> they were designed
> properly and that
> the bulk of users
> asking to change
> margins haven't even
> "read a bit on the
...
Just a random data point ... I could read this portion of
very narrow text somewhat faster than the other portions.
Maybe I didn't have to move my eyes side to side, or maybe
it was something else.
> Perhaps I will at some point. It might help me gain some insight into the
> minds of the LaTeX crowd.
The layout of the default LaTeX document classes is so terrible,
it gives me no confidence that any of its layout parameters
(such as line width) are good.
Donald Arseneau as...@triumf.ca
Rowland McDonnell
Yes they were - but... The problem is that the design isn't very good.
The process was fine; the outcome less so. The designer only paid
attention to *most* of the bits he thought he needed to as far as I can
tell, in order to create designs suitable for 1980s American academics.
The designs don't take into account whether or not you are in fact using
a header and/or footer line, for example. They don't take into account
the actual width of the fount in use - although I think that a
documentclass ought to be designed with a particular fount set in mind
so this might not be such a bad thing. There are lots of other problems
- all of which can be addressed with add-ons, such as switching to
KOMA-script or any number of bolt-on modifications such as my rmpage
package (re-write needed, and the docs are unfinished, but it's on CTAN
and I use it for everything).
[snip]
> > (one problem is that I'm pretty sure that most people *can't* run three
> > aesthetic senses simultaneously. I suspect that good typographers can
> > switch between the three views very rapidly. I find that I have to take
> > an overnight break to get sufficient distance from my previous outlook
> > ...
>
> This is why certain styles have taken me several hours to create, of
> course, and several hours to correct and/or improve, and hours to monitor
> and alter, and hours and hours and hours.
I measure the process in days, often longer. But that's because I'm
having to learn on the fly.
> My overnight break tends to
> occur as a matter of fact, simply because I'd pass out before projects
> were finished if I were trying to get them all done at once.
>
> > Having said that, you might find it useful to read the documentation for
> > the memoir class - it's not a full guide to typography or anything, but
> > it's got a lot of very interesting and useful things to say about how to
> > arrange text on your pages. ...
>
> Perhaps I will at some point. It might help me gain some insight into the
> minds of the LaTeX crowd.
>
> >> It gets old after a while to think that all these people asking these
> >> questions over and over and over again really people with doctorates
> >> in graphics design or background in typography.
>
> > I wouldn't trust anyone with a graphics design background myself.
>
> They have `authority', you know, those people with degrees in graphics
> design, etc.; at the least, they think they do.
They don't have any authority in my book. I'm happy with a graphics
designer laying out a page to a pre-planned style, but I'm not happy
with 'em coming up with the design in the first place. Erm. I have
worked as a professional magazine editor in that fashion: we had a
graphical designer moonlighting as our layout artist, but our magazine
styles had been designed by someone more expensive.
> But I suspect that most
> users I see asking questions have gotten their authority from Word: "It
> does it that way, so it must be right!"
>
> >> Alas, the drag-and-clickity interface of you-know-what has left people
> >> with this idea that they can do whatever they want. I guess they can;
> >> I guess that's why nothing looks good.
>
> > The problem is more that they are *ignorant*. I used WPs and then LaTeX
> > for *years* before starting to learn about typography ...
>
> This is true; a good number of people have a lot that they just don't care
> about.
>
> > But what can one do? I've taught typing and WP use professionally to
> > official exam standards ... There's lot about making sure you've got 1
> > inch margins in some of the standard textbooks - as if we were still
> > using typewriters on foolscap, fer gawd's sake.
>
> Yeah, and here at a university I've received a rather cold reception on
> the TeX/LaTeX front. It is my understanding that universities are
> supposed to be great havens for happy software such as mentioned, and for
> collections of *nix operating systems. I guess the problem at this
> particular university is that there are people and buildings named `Gates'
> and Redmond is 40 miles away.
The problem is more likely to be that `suits' want to use MS no matter
what and cannot trust anything that's free.
> > There's no hope until this sort of thing changes: WP publishers should
> > be producing style guides, and brief introductions to typography ...
> > And the people who design education and training for those in need of
> > typing instruction ought to add a touch on basic typgraphy ...
>
> Blech, WP styles.
WP styles can be a good thing. But I'm not talking about necessarily
teaching everyone about structured document preparation: just the basic
points you need to pay attention to in order to create decent documents.
> Well, I would suspect that they didn't do it before
> because it would have required reading a 2000 page manual. Then again,
> Computers and Typography is 2000 pages, so I'm not sure what that means.
I was thinking in terms of a couple of pages of notes and then a simple
tutorial on `how to set things up sensibly and *why this is sensible*' -
the latter is almost always missing in docs for all modern consumer
technology, not just computers.
> I think the majority of people who do the training probably don't know any
> better either, so it's likely a self-propogating problem.
I *know* that virtually no-one who does that sort of training has the
faintest idea in England at least (education works differently in
Scotland and NI, and who can tell what the Welsh get up to?)
[snip]
> >> I will just note one study that found that screen reading at 100
> >> characters per line was a bit faster than everyone's happy 75[1]. By
> >> extension, it is reasonable to read at wider widths than that. :)
>
> > Screen reading is significantly different to paper reading. I suspect
> > that it's harder to whizz back to the start of a line on a screen than
> > on paper - which might mean that excessively long lines don't introduce
> > a slow-down because you're suffering one anyway.
>
> I think when one has a cursor that it is quite easy to track the
> beginnings of the lines, hence this whizz-back is not a problem.
Er? What? I want to whizz back in a fraction of a second using my eyes
alone. I don't want to follow the text with my finger or a cursor on a
screen, which is even slower and more awkward than a finger.
> What the
> report showed was that reading at 100 characters was faster (statistically
> so, if memory serves).
Yes, but with what population, under what conditions, performing what
test exactly?
> This was amusing to me, since there are certain
> instances where I feel I'm reading much faster at 150 characters wide.
I've timed myself. I'm slower with blocks of text that have very long
lines than with sensible lines - but once you get beyond a certain
width, `the wider the better'.
[snip]
> > Another problem is the use of Computer Modern Bold Extended for
> > sectional headers, and the `feel' is very - well, 1980s USA, which was
> > fine for the 1980s Americans who came up with 'em but grates on me at
> > least.
>
> Perhaps this is what is really happening to me. I wasn't doing anything
> typographic in the 80s, so I wouldn't have noticed. :\
I don't understand what relevance that has. You can read 1980s print
publications to see the sort of thing I'm talking about. It's what I've
done.
[snip]
> To be quite frank about it, I find that TeX is very useful, but maybe I
> should still read some of the documentation and design decisions that were
> made for memoir and KOMA.
Highly recommended.
> I seem to recall noting rmpage at one point, when I was searching for
> something, so it has likely helped a good collection of people. I don't
> recall exactly what it does,
rmpage is a general purpose package for changing LaTeX page layout
parameters in a very different way to geometry: you tell it things like
`longer', `wider', noheaders, footers, and suchlike to permutate what
you get until you like it. It has many bells and whistles to do things
like ensure that your chosen page layout can be printed by your chosen
printer, ensure sensible layouts when you change to a different fount,
set up defaults and local mods in a config file, and so on. The design
is, erm, best described as `heavily baroque, possibly by a 3 year old'.
It really needs re-writing and the docs aren't finished yet - but as I
say, I use it for pretty much everything. Well, I wrote it to suit my
way of working so I would use it all the time, wouldn't I?
(rmpage isn't bad, just not what it ought to be and I've learnt a *lot*
since I started on it)
> but I think it was mentioned at another spot
> in this thread, and I heard a mention of lowercase widths and
> characters-per-line computations. I've been known to have TeX tell me
> this type of stuff
Probably not in the way I did in rmpage. I do it with Homer and a goat
sacrifice (*REALLY* - read rmpage for the shocking truth).
> so, if this is indeed the type of thing it does, you
> and I are of like mind (we certainly seem to be alike in our propensity
> for verbosity).
I hope you're slightly less foul-mouthed than me. I `do' many styles of
communication.
[snip]
> >> The table in Bringhurst [2] uses lowercase alphabet length, and seems
> >> to prefer columns with 60 to 69 characters.
> >>
> >> [1] "Effects of Line Lengths and Method of Movement on Patterns of Reading
> >> From Screen". Visible Language 32.2. Page 166.
> >> http://www.id.iit.edu/visiblelanguage/
> >>
> >> [2] Bringhurst, Robert. The Elements of Typographic Style. Version 2.4.
> >> Hartley & Marks, Publishers. Vancouver. 2001. p. 29.
>
> > Thank you for this.
>
> Alas, there are still people who don't know Bringhurst, so it is necessary
> to include the reference.
Umm. It's only two years old and a Canadian publication on a specialist
subject which not many people do research into. Not surprising most
people have never heard of it.
> Maybe it's not as bad as I think, but perhaps it's time I stop writing to
> newsgroups when I'm in a bad mood.
<wild and manic laughter>
Being in a bad mood doesn't seem to have stopped anyone else from
posting here, or haven't you been reading ctt long? ;-)
Rowland McDonnell
> Brian Blackmore <bl...@po.cwru.edu> writes:
>
> > My point was that
> > they were designed
> > properly and that
> > the bulk of users
> > asking to change
> > margins haven't even
> > "read a bit on the
> ...
> Just a random data point ... I could read this portion of
> very narrow text somewhat faster than the other portions.
> Maybe I didn't have to move my eyes side to side, or maybe
> it was something else.
One of the standard `tricks' of so-called speed-reading is to avoid
side-to-side eye movement and simply scan downwards - I expect you were
benefitting from that. It's possible to learn to read at astonishing
speed (I know someone who has demonstrated this to me - she mocked for
being such a slow reader because I wasn't keeping up even slightly.
This surprised me because I'm a pretty damned rapid reader in the
conventional sense. Then she admitted she'd taken the trouble to learn
speed reading...) if you get the hang of it - but the more sideways
scanning you do, the slower you are. I've been told that the way to
learn how to speed read is to start by reading the lines in a couple of
jumps rather than a continuous flow, and build up from there.
[snip]
Lucian Wischik
>One of the standard `tricks' of so-called speed-reading is to avoid
>side-to-side eye movement and simply scan downwards - I expect you were
>benefitting from that. It's possible to learn to read at astonishing
>speed (I know someone who has demonstrated this to me - she mocked for
>being such a slow reader because I wasn't keeping up even slightly.
>This surprised me because I'm a pretty damned rapid reader in the
>conventional sense. Then she admitted she'd taken the trouble to learn
>speed reading...)
I never understood that. Why do you have to learn it? Why doesn't it
come naturally? Speed-reading is my natural way of reading which I've
used as long as I can remember, and it's hard for me not to do it.
Makes it very hard to read poems... but helps for reading Usenet!
--
Lucian
Jellby
>> > My point was that
>> > they were designed
>> > properly and that
>> > the bulk of users
>> > asking to change
>> > margins haven't even
>> > "read a bit on the
>> ...
>> Just a random data point ... I could read this portion of
>> very narrow text somewhat faster than the other portions.
>> Maybe I didn't have to move my eyes side to side, or maybe
>> it was something else.
>
> One of the standard `tricks' of so-called speed-reading is to avoid
> side-to-side eye movement and simply scan downwards - I expect you were
> benefitting from that.
Would it be faster to read a 'vertical' language, like Chinese, Japanese,
Korean or Mongolian? I guess not, because you still have to scan the 2
dimensions of the writing... unless you apply the speed-reading and avoid
the vertical eye movement ;)
--
Ignacio __ Fernández Galván
/ /\
Linux user / / \
#289967 / / /\ \ PGP Pub Key
/ / /\ \ \ 0x01A95F99
/ /_/__\ \ \
/________\ \ \
\___________\/
Brian Blackmore
> Brian Blackmore <bl...@po.cwru.edu> wrote:
>> Rowland McDonnell <real-addr...@flur.bltigibbet> wrote:
[chomp]
>> >> The table in Bringhurst [2] uses lowercase alphabet length, and seems
>> >> to prefer columns with 60 to 69 characters.
>> >>
>> >> [1] "Effects of Line Lengths and Method of Movement on Patterns of
>> >> Reading From Screen". Visible Language 32.2. Page 166.
>> >> http://www.id.iit.edu/visiblelanguage/
>> >>
>> >> [2] Bringhurst, Robert. The Elements of Typographic Style.
>> >> Version 2.4. Hartley & Marks, Publishers.
>> >> Vancouver. 2001. p. 29.
>>
>> > Thank you for this.
>>
>> Alas, there are still people who don't know Bringhurst, so it is necessary
>> to include the reference.
> Umm. It's only two years old and a Canadian publication on a specialist
> subject which not many people do research into. Not surprising most
> people have never heard of it.
The copyright list is 1992, 1996, 2001, so the 2001 copyright is for
Version 2.4. It's, you know, "The Typographers Bible" according to a
number of sources.
Rowland McDonnell
> real-addr...@flur.bltigibbet (Rowland McDonnell) wrote:
> >One of the standard `tricks' of so-called speed-reading is to avoid
> >side-to-side eye movement and simply scan downwards - I expect you were
> >benefitting from that. It's possible to learn to read at astonishing
> >speed (I know someone who has demonstrated this to me - she mocked for
> >being such a slow reader because I wasn't keeping up even slightly.
> >This surprised me because I'm a pretty damned rapid reader in the
> >conventional sense. Then she admitted she'd taken the trouble to learn
> >speed reading...)
>
> I never understood that. Why do you have to learn it?
Because no-one is born able to read. Reading is one of the things like
walking and talking that one has to learn. Breathing, eating (well, the
first sort of feeding you do), crying, puking, shitting, and suchlike
don't need to be learnt - pretty much everything else does.
> Why doesn't it
> come naturally?
I think that it probably *would* come naturally if one were taught to
read in a different way.
> Speed-reading is my natural way of reading which I've
> used as long as I can remember, and it's hard for me not to do it.
Lucky you. I believed the teachers who taught me to read. Conventional
reading instruction teaches you to do the job in a strict raster scan
(what did I just say?) one letter at a time and then one word at a time.
If you don't think that any other approach can exist, why would one look
for one or try one?
I can't help feeling, since the younger you are the better you are at
learning, and that the brain goes through two separate major structural
changes when you're growing up, that if you try to learn to speed-read
after the age at which you ought to be reading - your very early school
years - it's harder to learn and you can't get as good at it.
There's a big brain change when you're about 3ish - which is why almost
no-one can remember anything before about that age; and there's another
big brain change while you're adolescentish - which is in part why
teenagers are such a pain. Before the last big brain change that's
known about in human development, a human being is typically pretty bad
at predicting consequences and acting sensibly on those predictions (I
don't know why, because I was probably better at verbal reasoning when I
was young than now, so I *could* do the required analysis; it's just
that I didn't think to do so). If teenagers seem inconsiderate of
others, that's because they are but they can't help it and they're
developing brains that will allow them to be properly considerate. It's
not that they don't care, it's just that before you get `grown up', you
just can't think like an adult about the full consequences of what
you're doing. btw, another effect of what's going on then is to require
the growing person to have a lot more sleep than normal *AND* to live
slightly out of phase with the rest of humanity - getting up late and
going to bed late. Yes, all that is in fact caused by physiological
changes - your annoying teenaged child who seems to do nothing but
sleep, eat you out of house and home, and snarl incoherently you while
being capable of lively and interesting conversations lasting all night
on your telephone line, is going through a very traumatic experience and
frankly can't be expected to be decent all the time - so there's only
really justification for breaking *one* leg[1] you finally crack. The
speculation I've heard about why evolution gave us this setup is that
there's probably some sort of social benefit in having these annoying
creatures only associating with each other until they've got sensible,
and also to allow them to develop new ideas and whatnot slightly
detached from the older generation.
(the perils of modern teacher training...)
> Makes it very hard to read poems... but helps for reading Usenet!
Yep. I think I'll probably take the time to learn and practice myself,
`real soon now' as they say...
Rowland.
[1] I know, no decent parent would do that. After all, you'd have the
trouble of taking 'em to hospital and then have to put up with 6 weeks
of clunking around the place in a plaster cast[2] and needing even more
lifts in the car than usual.
[2] Or, in the case of my little bro, having to have a wheelchair[3] -
which is great when he's five and you're not much older, 'cos it's small
enough to be a wonderful indoor toy when your parents have a relaxed
attitude towards scuffed paintwork and lots of nice polished wooden
floors.
[3] Got run over by a motorcycle - on the school crossing after having
been waved across by the lollipop lady - broke an arm and a leg. His
luck tends to be like that. The bike rider was, you'll be pleased to
hear, utterly mortified. The lollipop lady kept her job for some
reason.
Rowland McDonnell
> Among other things, Rowland McDonnell wrote:
[snip]
> > One of the standard `tricks' of so-called speed-reading is to avoid
> > side-to-side eye movement and simply scan downwards - I expect you were
> > benefitting from that.
>
> Would it be faster to read a 'vertical' language, like Chinese, Japanese,
> Korean or Mongolian? I guess not, because you still have to scan the 2
> dimensions of the writing... unless you apply the speed-reading and avoid
> the vertical eye movement ;)
I don't know much about Oriental languages. But I was under the
impression that non-alphabetical scripts like Chinese can be written to
be read in any direction.
Rowland McDonnell
> Rowland McDonnell <real-addr...@flur.bltigibbet> wrote:
> > Brian Blackmore <bl...@po.cwru.edu> wrote:
> >> Rowland McDonnell <real-addr...@flur.bltigibbet> wrote:
> >> > Brian Blackmore <bl...@po.cwru.edu> wrote:
[snip]
> >> >> [2] Bringhurst, Robert. The Elements of Typographic Style.
> >> >> Version 2.4. Hartley & Marks, Publishers.
> >> >> Vancouver. 2001. p. 29.
> >>
> >> > Thank you for this.
> >>
> >> Alas, there are still people who don't know Bringhurst, so it is necessary
> >> to include the reference.
>
> > Umm. It's only two years old and a Canadian publication on a specialist
> > subject which not many people do research into. Not surprising most
> > people have never heard of it.
>
> The copyright list is 1992, 1996, 2001, so the 2001 copyright is for
> Version 2.4. It's, you know, "The Typographers Bible" according to a
> number of sources.
Righto. btw, I started to look at books on typography when the *first*
edition of that book was 2 or 3 years old - there were no references to
it that I came across back then. Mind you, I was having to use the
local public reference library at the time. A decent place with helpful
librarians, but - well, it's run by local government in England so see
if you can guess how well funded it was.
Danie Els
> Danie Els <dnj...@netscape.net> wrote:
>
[snip]
> > hold the book (for your thumbs). One of the main reasons for the wide
> > margins is that the book looks good as a whole. To quote Bringhurst: "
> > Perhaps fifty per cent of the character and integrity of a printed
> > page lies in its letterforms. Much of the other fifty per cent resides
> > in its margins."
>
> There's a lot of rubbish talked about typography.
I don't think Bringhurst is a Micky Mouse in the typography world. He
is regarded very highly by people such as Herman Zapf whom states on
the back cover of the book that he wish to see this book become the
Typographers Bible.
> > The one thing I don't like about the LaTeX page layout is the fact
> > that the textwidth is fix to plus-minus 75 character of Computer
> > Modern font for all font sizes.
>
> That's not the LaTeX page layout; it's just what you get with the
> standard LaTeX classes. [snip]
Granted, yes its only in book, report, article, letter, amsbook,
amsart, memoir and nearly every other LaTeX class ;-)
[snip]
> > For my own purposes (A4 pages) I
> > normly fix the text width and height and then select the font size to
> > give a decent number of character.
>
> <curious> Why not do your fount selection as an integral part of your
> document design? That is, say `I want a wide measure in one column, so
> I'd better pick a wide setting typeface', sort of thang.
I hate this setups that say you must have 66.0000001 characters on a
line. It depends very much on the font . Some can take more, others
less. Take for example Palatino and Aldus. Palatino is a display font
while Aldus is it's text variant. I personally feels that Aldus can
take a lot more characters per line than Palatino.
I still feels that the fixed margins works better for A4 paper. If you
are talking of different paper sizes, it is a different story.
> > On a A4 page, if you select a type block with width 2/3 page width and
> > the height to width ration equal to the golden ratio, you find quite a
> > pleasing layout (better than the latex one).
>
> Some argue that decoupling the text body aspect ratio from the
> surrounding paper aspect ratio, that you get a non-optimal result.
Sorry, but I think that was one of the contributions of the typewriter
and later MS-word towards typographic excelence.
> Others, obviously, disagree. Me? I have no fixed opinion, but also
> often type `ration' instead of `ratio'.
>
A4 paper is not very typgraphic friendly and although the 1:2^0.5 page
ratio is known for hundreds or thounds of years, yet the only decent
textblock ratio for it is still the golden ratio.
[snip]
>
> ppl 12pt is 5.26640 pt/char, giving 75.6 chars across 2/3 of A4
> cmr 12pt is 5.16339 pt/char, giving 77.1 chars across 2/3 of A4
> ptm 12pt is 4.76569 pt/char, giving 83.6 chars across 2/3 of A4
>
I have a macro based on a least squares fit of the Bringhurst (sigh)
typesetting tabel (there is something simmilar in memoir). I would
like to test it against the rmpage algorithm. So, some questions: Does
the count 1067 characters in the Homer text includes spaces and
punctuation? The setting in a long horisontal box does not contain
any stretch between the words, so real text would actualy contain less
character in a line than the theoretical value. Did you have a look at
this effect?
[snip]
>
> btw, rather than saying `artistic', I'd say `aesthetic'; but that's
> because I'm with Oscar Wilde in thinking that all art is by definition
> useless, and we don't want useless page layouts, do we?
>
Ja, that was more or less what I was trying to say, but as a
non-native English speaker I am often without words (ratio, ration,
whatever). In Afrikaans we have a saying: English is not my first
language, it is not even my last language.
Regading the wisdom of Oscar Wilde, damn that guy was good!
David Stone
> The thing about these margins that you see is that they *are* an
> appropriate size. The problem is the paper size, not LaTeX's
> typesetting.
>
> ... A4 paper is just too wide for single columns of text in
> 10pt-12pt sizes. It's great for old-fashioned typewriting (take a
> look at the sheer width of typescript), but rotten for proper
> printing founts - because you get margins that are just too wide, as
> you've noticed. The problem is that you shouldn't reduce the size
> of the margins, because that'll make the text awkward to read even
> though the text looks better on the page from a distance.
>
> The only sane way round this is to use two columns, I reckon[1].
assuming we're stuck with a4 paper, yes. Given that nobody uses
typewriters any more, it seems odd that we still all use a4 (or
u.s. letter, which is slightly worse). I tried another way round the
problem: buy and use b5 paper, but I could not find anywhere that
would sell it in the u.k. (It seems to be available in Japan, but
that's not very convenient for me.) If anybody reading this knows
where to buy b5 paper, it would be appreciated if they'd say where.
--
David Stone
Jesper Harder
> The problem is that A4 paper is just too wide for single columns of
> text in 10pt-12pt sizes. [..] The problem is that you shouldn't
> reduce the size of the margins, because that'll make the text
> awkward to read even though the text looks better on the page from a
> distance.
>
> The only sane way round this is to use two columns, I reckon[1].
The problem with two columns is that it's too narrow to accommodate
long equations.
If you look at a typical Physics journal in two column layout (say,
phys.rev.a) you'll often see that some parts the page is one column,
and even some pages which are entirely one column, because fitting the
equations in one column was too difficult.
It's an awful design -- ugly and confusing.
Rowland McDonnell
> real-addr...@flur.bltigibbet (Rowland McDonnell) writes:
[snip]
> > The only sane way round this is to use two columns, I reckon[1].
>
> assuming we're stuck with a4 paper, yes. Given that nobody uses
> typewriters any more, it seems odd that we still all use a4 (or
> u.s. letter, which is slightly worse).
Nah. Think filing cabinets.
[snip]
Rowland.
Rowland McDonnell
> real-addr...@flur.bltigibbet (Rowland McDonnell) wrote:
> > Danie Els <dnj...@netscape.net> wrote:
> >
> [snip]
> > > hold the book (for your thumbs). One of the main reasons for the wide
> > > margins is that the book looks good as a whole. To quote Bringhurst: "
> > > Perhaps fifty per cent of the character and integrity of a printed
> > > page lies in its letterforms. Much of the other fifty per cent resides
> > > in its margins."
> >
> > There's a lot of rubbish talked about typography.
>
> I don't think Bringhurst is a Micky Mouse in the typography world.
Maybe he's not. But Albert Einstein wasn't a Mickey Mouse either; and
he went to his grave denouncing quantum mechanics (in part an invention
of his) as a load of old rubbish - `God does not play dice with the
universe' and all that. Lord Kelvin (aka William Thom(p?)son; an Irish
physicist from back when Ireland wasn't independent and one of the most
brilliant scientists of his age, but rather odd in that he apparently
couldn't understand any scientific idea unless it was presented in the
form of mathematics) wasn't a Mickey Mouse either - and stated, at the
end of the 19th century or thereabouts, that physics was basically all
sorted out with all the important mechanisms fully understood, and all
we needed was better measurements. One of his more stupid thoughts, it
has to be said.
The point I'm getting at is it doesn't matter how bright you are, how
well informed, or how well regarded you are: you are still perfectly
capable of coming out with utter bollocks, and probably will.
Physicists in particular do this sort of thing quite a lot.
(Although not as often as people think - the famous occasion on which a
physicist announced at a dinner party that bumblebees can't fly was
caused by him scribbing some very rough calculations on the back of a
napkin to try to answer the question `How do bees fly'. He concluded
that they couldn't - and announced this fact in an attempt to explain
that he couldn't answer the question, because the only calculations he
could do were far to approximate to be of any use, and he'd need
computing facilities and reference books to do better.)
In short, I don't take authority as a reason for believing someone: I
want to check the working. If someone well regarded in a field makes a
pronouncement, I want to check up on where it comes from. I don't care
if he's the Lord High Panjandrum of Whatever or even the Akond of
Swat[1]: if he's not providing the working, I treat the statement with
deep suspicion.
> He
> is regarded very highly by people such as Herman Zapf whom states on
> the back cover of the book that he wish to see this book become the
> Typographers Bible.
That doesn't mean it's anything but insane to take his word as holy writ
simply because it's his word.
> > > The one thing I don't like about the LaTeX page layout is the fact
> > > that the textwidth is fix to plus-minus 75 character of Computer
> > > Modern font for all font sizes.
> >
> > That's not the LaTeX page layout; it's just what you get with the
> > standard LaTeX classes. [snip]
>
> Granted, yes its only in book, report, article, letter, amsbook,
> amsart, memoir and nearly every other LaTeX class ;-)
So what? It's easy to change, and there are plenty of mechanisms for
doing so using a variety of bolt-on goodies, many of which come as a
standard fitting with LaTeX these days. And *not* changing \textwidth
from that value if you're doing two column work is frankly stupid.
> [snip]
> > > For my own purposes (A4 pages) I
> > > normly fix the text width and height and then select the font size to
> > > give a decent number of character.
> >
> > <curious> Why not do your fount selection as an integral part of your
> > document design? That is, say `I want a wide measure in one column, so
> > I'd better pick a wide setting typeface', sort of thang.
>
> I hate this setups that say you must have 66.0000001 characters on a
> line. It depends very much on the font . Some can take more, others
> less. Take for example Palatino and Aldus. Palatino is a display font
> while Aldus is it's text variant. I personally feels that Aldus can
> take a lot more characters per line than Palatino.
Well, yes - so surely you should design the page layout for a particular
fount, making that sort of compromise decison based on the paper size,
the number of columns, and the particular characteristics of the
particular fount you are using. I think that's much better than saying
`I want to use so much of the paper width' or `I want to use so many
characters'. I'd say `Well, this fount works in a narrow/wide column,
this paper's so big, here's the job I want to do, so I need something
about like this. Hmm. No, like *this*' (and so on). If you see what I
mean.
> I still feels that the fixed margins works better for A4 paper. If you
> are talking of different paper sizes, it is a different story.
I reckon it's best to be very flexible about the whole thing. That's in
part why I put the incremental adjustment ability into rmpage.
> > > On a A4 page, if you select a type block with width 2/3 page width and
> > > the height to width ration equal to the golden ratio, you find quite a
> > > pleasing layout (better than the latex one).
> >
> > Some argue that decoupling the text body aspect ratio from the
> > surrounding paper aspect ratio, that you get a non-optimal result.
>
> Sorry, but I think that was one of the contributions of the typewriter
> and later MS-word towards typographic excelence.
I know you're wrong: I know for a fact that it results in loads of
people using margins that are far too small because that's what the
defaults are and that's what they were taught to do.
[snip]
> > ppl 12pt is 5.26640 pt/char, giving 75.6 chars across 2/3 of A4
> > cmr 12pt is 5.16339 pt/char, giving 77.1 chars across 2/3 of A4
> > ptm 12pt is 4.76569 pt/char, giving 83.6 chars across 2/3 of A4
> >
>
> I have a macro based on a least squares fit of the Bringhurst (sigh)
> typesetting tabel (there is something simmilar in memoir). I would
> like to test it against the rmpage algorithm. So, some questions: Does
> the count 1067 characters in the Homer text includes spaces and
> punctuation?
Yes - btw, it's not just (translated) Homer.
> The setting in a long horisontal box does not contain
> any stretch between the words, so real text would actualy contain less
> character in a line than the theoretical value.
Certainly, but so what?
> Did you have a look at
> this effect?
No. I worked on the assumption that it didn't matter - all I really
needed was a good guide to the width of the fount which made some sort
of sense. The exact numbers don't matter: what matters is that the
numbers have a consistent relationship to the width of the founts in
question. I *assumed* this was the case, on the grounds that this sort
of purely numerical calculation is never going to be what's needed to do
the job: that requires human aesthetic judgment applied to the printed
output. Fount metrics are a very poor guide to this sort of thing. The
idea was to get something about right done, and assume that the human
user can do the rest.
> [snip]
> >
> > btw, rather than saying `artistic', I'd say `aesthetic'; but that's
> > because I'm with Oscar Wilde in thinking that all art is by definition
> > useless, and we don't want useless page layouts, do we?
> >
> Ja, that was more or less what I was trying to say, but as a
> non-native English speaker I am often without words (ratio, ration,
> whatever).
I read in the Jargon file that in German, for example, hacking terms are
integrated fully into the language - so a German might grep for
something as an English person. So they had to decline these invented
verbs - grep, greppen, gegrept, for example. Well, if these invented
words work in English, I don't suppose it's a surprise that they work in
a different Germanic language, is it?
> In Afrikaans we have a saying: English is not my first
> language, it is not even my last language.
<chuckle>
Why not call it your last resort? I reckon that's the best way of
looking at any language which is used as in a country due to some
reasonably recent colonial influence - Latin's too ancient for anyone to
view it like that, I reckon[2].
btw, P.J. O'Rourke described (in one of his many successful attempts to
get me to exchange hard cash for paper covered in ink) West Germans are
apparently identical to North Americans - clean, well-dressed in the
height of fashion, orthodontically corrected, up with the latest music
etc.; but there is one major difference according to P.J.: West Germans
are all fluent in English.
> Regading the wisdom of Oscar Wilde, damn that guy was good!
Well, *sometimes*. Suing the Marquis of Queensbury for libel wasn't a
bright idea - Wilde sued basically because the Marquis had called him a
poof. Unfortunately, Oscar Wilde *was* a poof... Not very clever,
Oscar - I mean, you must have noticed a few hints that you were
homosexual, Mr Wilde, like the way you kept going to bed with the
Marquis of Queensbury's son, Lord Alfred Douglas. The `thunk' you can
hear is my head hitting the wall in despair at human stupidity.
(I might be making light of the above episode, but that's only because
all concerned are long dead. The disastrous libel case, publicity, and
the fact that Oscar Wilde then got locked up (the `Ballad of Reading
Gaol' was about the only good result) because homosexuality was illegal
back then; all that ruined his life completely, and the family changed
its name to `Holland' (IIRC) to escape the link with Wilde - his
grandson Merlin Holland (IIRC) scribbles for a living as far as I can
tell)
Rowland.
[1] Mr Leakey is one of my friends too.
[2] I could entertain the idea of, for example, demanding an apology
from the Italian government for invading by country and enslaving my
ancestors back when the Romans invaded Britain. But since my ancestors
included the Romans who invaded and also the Britons who refused to have
any of this Roman nonsense at all, *and* a whole bunch of other people
who didn't have a damned thing to do with it in *this* part of the world
(and some people from much further east who had a history of running
their own campaigns of invasion and slaughter, probably quite
independent of Brutus coming to Britain fallowing the fall of Troy, and
that's just the things I can track down), it'd be a bit sillier than
that sort of nonsense normally is.