fixed line spacing with LaTeX?
Dr. Joel M. Hoffman
I'm typesetting a book with LaTeX, and I need a way to make sure that
lines are in the same place on every page. For example:
PAGE 1 PAGE 2
text text text text text text text text
text text text text
text text text text
HEADER text text text text
text text text text
text text text text
text text text text text of next para.
text text text text text text text text
text text text text text text text text
Notice that text lines up from page to page. Can LaTeX do this? The
problem, of course, is that my leading is 10/12pt, while some header
text is is 12 or even 14pt. If, for example, I include a 12pt header
surrounded by some space, I need a command to "go down to the nearest
12pt interval" before I start typesetting lines again.
Any suggestions?
Many thanks.
-Joel
p.s.: This seems to be a very basic requirement of modern
typesetting. I'm surprised TeX and LaTeX don't do this by default....
Johan
> p.s.: This seems to be a very basic requirement of modern
> typesetting. I'm surprised TeX and LaTeX don't do this by default....
Latex allows--by design--various inter-line spacings to stretch
somewhat, to avoid oceans of white on the page.
To override this behavior, you'll need to find the various s.c. lengths
involved and set them to absolute values, rather than stretchy ones.
10pt is a fixed lenght, while 1cm plus 3cm minus 2cm is a "rubber
length". Do a google news search for "tex rubber length plus minus"; the
mar 20, 2000 discussion explains it fairly well.
I imagine you'll need to do some arithmetic to make Section headings and
such have heights that are an integral number of baselineskips.
The tradeoff is that you'll restrict how good tex will be able to make
your document look.
Perhaps all you need is to lock certain lines in place without
constraining the stretch of others. I've wanted to do this myself in the
horizontal direction: somehow bind the "current cursor x-offset from
left margin" to a variable, and also "advance cursor to absolute
x-offset from margin".
Can anyone suggest a way to do that?
Jellby
> Latex allows--by design--various inter-line spacings to stretch
> somewhat, to avoid oceans of white on the page.
>
> To override this behavior, you'll need to find the various s.c. lengths
> involved and set them to absolute values, rather than stretchy ones.
As a side-question, is it possible to specify a "constant" line spacing in
text? I mean a line spacing that's not affected by an inline formula with
too large height/depth. I'd rather have a subscript partly overlapping an
ascender than a larger spacing between two lines of text. Or, at least,
have the same larger spacing for the whole paragraph.
I guess I could use setspace (to enlarge line spacing where needed) or
makebox (to make the too-high formulas look like not-so-high to latex). But
is there a more automatic way to get this constant line spacing?
--
Ignacio __ Fernández Galván
/ /\
Linux user / / \
#289967 / / /\ \ PGP Pub Key
/ / /\ \ \ 0x01A95F99
/ /_/__\ \ \
/________\ \ \
\___________\/
David Kastrup
> Among other things, Johan wrote:
>
>
> > Latex allows--by design--various inter-line spacings to stretch
> > somewhat, to avoid oceans of white on the page.
> >
> > To override this behavior, you'll need to find the various s.c. lengths
> > involved and set them to absolute values, rather than stretchy ones.
>
> As a side-question, is it possible to specify a "constant" line
> spacing in text? I mean a line spacing that's not affected by an
> inline formula with too large height/depth. I'd rather have a
> subscript partly overlapping an ascender than a larger spacing
> between two lines of text. Or, at least, have the same larger
> spacing for the whole paragraph.
>
> I guess I could use setspace (to enlarge line spacing where needed)
> or makebox (to make the too-high formulas look like not-so-high to
> latex). But is there a more automatic way to get this constant line
> spacing?
The experimental output routine from the LaTeX3 code base offers that.
--
David Kastrup, Kriemhildstr. 15, 44793 Bochum
UKTUG FAQ: <URL:http://www.tex.ac.uk/cgi-bin/texfaq2html>
Robin Fairbairns
>I'm typesetting a book with LaTeX, and I need a way to make sure that
>lines are in the same place on every page. For example:
frank mittelbach presented a paper at tug2000 about a new output
routine for latex, which (almost as a throw-away remark) provides grid
spacing of lines. i typeset the paper for tugboat with the revised
output routine, which was a bit of a strain, but you can see the
results at:
http://www.tug.org/TUGboat/Articles/tb21-3/tb68mittel.pdf
i don't believe it's quite ready for the big time, yet, but it plainly
_is_ possible. the output routine (xor, if my brain hasn' faded
further than usual), can be found on www.latex-project.org somewhere.
--
Robin (http://www.tex.ac.uk/faq) Fairbairns, Cambridge
Dr. Joel M. Hoffman
>> typesetting. I'm surprised TeX and LaTeX don't do this by default....
>
>Latex allows--by design--various inter-line spacings to stretch
>somewhat, to avoid oceans of white on the page.
Right. The fundamental problem with LaTeX that I'm facing is that,
while it does a great job a page layout, it does a terrible job of
multi-page layout, having, for example, no notion of "spread" (facing
pages, the bottoms of which should match) or back-to-back pages, the
lines on which should match, so that the lines on the page match the
lines one sees through the page.
>10pt is a fixed lenght, while 1cm plus 3cm minus 2cm is a "rubber
>length". Do a google news search for "tex rubber length plus minus"; the
> mar 20, 2000 discussion explains it fairly well.
This doesn't help with Sections, in-line floats, etc.
>The tradeoff is that you'll restrict how good tex will be able to make
>your document look.
No. I'll restrict how good each page can look, but improve the
document. At least, that's what I'm learning.
At any rate, does anyone know how to do this?
Thanks!
-Joel
Dr. Joel M. Hoffman
>>length". Do a google news search for "tex rubber length plus minus"; the
>> mar 20, 2000 discussion explains it fairly well.
>
>This doesn't help with Sections, in-line floats, etc.
>
Or lists, for that matter. Basically, I don't want to spend the next
two days looking through every bit of glue that ever gets put on the
page, and changing it to a multiple of 12pt. And even if I did, I'd
have to look at anything that uses type that's larger than 10pt, and
adjust the spacing around it.
But, seeing as no one seems to have a general solution to this
problem, can someone remind me how I know how far down on the page I
am so far? Then I can module 12pt it, and add the right amount to
bring me back to a standard 12pt baseline.
Thanks.
-Joel
Donald Arseneau
> >>10pt is a fixed lenght, while 1cm plus 3cm minus 2cm is a "rubber
> >>length". Do a google news search for "tex rubber length plus minus"; the
> >> mar 20, 2000 discussion explains it fairly well.
> >
> >This doesn't help with Sections, in-line floats, etc.
>
> Or lists, for that matter. Basically, I don't want to spend the next
> two days looking through every bit of glue that ever gets put on the
> page, and changing it to a multiple of 12pt.
Well, unless you can score such a document class from someone
else, that is exactly what you have to to. Mind you, how hard can
it be to search the document class file for the words "plus" and
"minus". If that takes you two days, I hope you get paid by the
hour.
Donald Arseneau as...@triumf.ca
Brooks Moses
This still has the difficulty that it only fixes part of the problem.
If there is a float at the top of the page, that float (and the margins
around it) would need to be adjusted to line the text up properly, as
well -- which would seem to me to be much more difficult than grepping a
class file for "plus" and "minus".
- Brooks
--
The "bmoses-nospam" address is valid; no unmunging needed.
Dr. Joel M. Hoffman
>If there is a float at the top of the page, that float (and the margins
>around it) would need to be adjusted to line the text up properly, as
>well -- which would seem to me to be much more difficult than grepping a
>class file for "plus" and "minus".
Right. The trick is that some things (interlinespace, e.g.) have to
have no glue, but other things *have* to have glue to make up for it.
One terrible way to do this would be to enclose every float, list, eps
file, etc., in a "make me a multiple of 12pt" vbox, which would set
the item at its natural height, measure it, and add enough space to
make it an even multiple of 12pt. If I do this, I'll have to have
strechable glue at the end of every such vbox, and, actually,
strechable glue inside it, too, would make the float look much better.
One thing the specs call for is 9-pt text in tables, which means the
tables won't line up, but they have to end at a multiple of 12pt.
Glue just at the end won't look nearly as good as glue throughout the
table.
My current thinking is to use \everypar, and find a way of calculating
the current place on the page, and then add enough space to start each
para at a multiple of 12pt. But I don't know (yet) how to figure out
the current place on the page. I have a vague recollection that
there's a parameter "output height so far" but I fear that's in TeX's
output routines, not LaTeX's.
-Joel
Patrick TJ McPhee
Dr. Joel M. Hoffman <jo...@exc.com> wrote:
% My current thinking is to use \everypar, and find a way of calculating
% the current place on the page, and then add enough space to start each
% para at a multiple of 12pt.
One problem with this is that the current place on the page is not
necessarily known at the time the paragraph is typeset. The TeX
primitive which gives the current position, so far as TeX knows, is
\pagetotal. One problem with using this in the way you want is that
it doesn't take into account glue, so for instance, given this
input
The next paragraph shows the page total.
\vfil
Here it is: \the\pagetotal.
\eject
The next paragraph shows the page total.
Here it is: \the\pagetotal.
\bye
Both second paragraphs read
Here it is 11.94444pt.
although one of them is roughly 24 pt from the top of the page, and the
other is right at the bottom of the page.
--
Patrick TJ McPhee
East York Canada
pt...@interlog.com
Dr. Joel M. Hoffman
>% the current place on the page, and then add enough space to start each
>% para at a multiple of 12pt.
>
>One problem with this is that the current place on the page is not
>necessarily known at the time the paragraph is typeset. The TeX
>primitive which gives the current position, so far as TeX knows, is
>\pagetotal. One problem with using this in the way you want is that
>it doesn't take into account glue, so for instance, given this
>input
So I guess the trick is not to have any unaccoutned-for glue. If I
take care only to add glue where I know it's being added, and only add
it such that after the glue the text is back to a multiple of 12pt,
that should work, right? I have in mind something like:
\amounttoadd=12pt - (\pagetotal mod 12pt)
\skip\amounttoadd
-Joel
Giuseppe Bilotta
It looks like slowly but surely LaTeX3 is heading towards where
ConTeXt has already been for a long time ;) (Ok, ConTeXt never had to
deal with centuries-old legacies ...)
--
Giuseppe "Oblomov" Bilotta
Can't you see
It all makes perfect sense
Expressed in dollar and cents
Pounds shillings and pence
(Roger Waters)
Rolf Marvin Bře Lindgren
> As a side-question, is it possible to specify a "constant" line spacing in
> text? I mean a line spacing that's not affected by an inline formula with
> too large height/depth. I'd rather have a subscript partly overlapping an
> ascender than a larger spacing between two lines of text. Or, at least,
> have the same larger spacing for the whole paragraph.
as a side-question, why would you want it? typesetting to a grid was
forced upon the typographers of old because of the equipment they used,
and it looks nice; but where is the evidence that forced spacing
increases legibility or readability?
--
Rolf Lindgren http://www.roffe.com/
ro...@tag.uio.no
David Kastrup
> It looks like slowly but surely LaTeX3 is heading towards where
> ConTeXt has already been for a long time ;) (Ok, ConTeXt never had
> to deal with centuries-old legacies ...)
Like amsmath? Like all of the various table creating packages? Like
multiple footnote support (to which I'll be adding my own
contribution soon)? Like literally thousands of packages available
for a variety of tasks?
While I do consider the _core_ of LaTeX suitable for replacing with
something more versatile, ConTeXt is quite far behind LaTeX with
respect to available addons.
Brooks Moses
Some people consider aesthetic appeal (aka "looks nice") to be
sufficient reason.
In this case, it's specifically the aesthetics of having text look
relatively uniformly spaced, rather than having big white lines through
it where a linespacing was increased to fit a tall formula.
One possible way to accomplish it would be to enclose all of one's
inline formulas in boxes of a sort that "hide" the extra height from the
paragraph formatting routines. I remember seeing something about how to
do that in Lamport's book, but I don't remember the details and it isn't
handy.
Scott Pakin
> One possible way to accomplish it would be to enclose all of one's
> inline formulas in boxes of a sort that "hide" the extra height from the
> paragraph formatting routines. I remember seeing something about how to
> do that in Lamport's book, but I don't remember the details and it isn't
> handy.
Perhaps you're thinking of the optional arguments to \raisebox, which
specify a height and depth to fake:
Blah, blah, blobbidy blah, blah, blah.
Blah, blah, blobbidy blah, blah, blah.
Blah, blah, blobbidy blah, blah, blah.
Blah, blah, blobbidy blah, blah, blah.
Blah, blah, blobbidy blah, blah, blah.
This is a test of
\raisebox{0pt}[0pt][0pt]{$\displaystyle\int_0^\infty \frac{1}{x^2}\,dx$}
and other such things.
Blah, blah, blobbidy blah, blah, blah.
Blah, blah, blobbidy blah, blah, blah.
Blah, blah, blobbidy blah, blah, blah.
Blah, blah, blobbidy blah, blah, blah.
Blah, blah, blobbidy blah, blah, blah.
-- Scott
Brooks Moses
> Brooks Moses wrote:
> > One possible way to accomplish it would be to enclose all of one's
> > inline formulas in boxes of a sort that "hide" the extra height from the
> > paragraph formatting routines. I remember seeing something about how to
> > do that in Lamport's book, but I don't remember the details and it isn't
> > handy.
>
> Perhaps you're thinking of the optional arguments to \raisebox, which
> specify a height and depth to fake:
>
> Blah, blah, blobbidy blah, blah, blah.
> \raisebox{0pt}[0pt][0pt]{$\displaystyle\int_0^\infty \frac{1}{x^2}\,dx$}
> and other such things.
> Blah, blah, blobbidy blah, blah, blah.
Yes, I'm pretty sure that's exactly what I was thinking of. Thanks!
Giuseppe Bilotta
> Giuseppe Bilotta <bilo...@hotpop.com> writes:
>
> > It looks like slowly but surely LaTeX3 is heading towards where
> > ConTeXt has already been for a long time ;) (Ok, ConTeXt never had
> > to deal with centuries-old legacies ...)
>
> Like amsmath? Like all of the various table creating packages? Like
> multiple footnote support (to which I'll be adding my own
> contribution soon)? Like literally thousands of packages available
> for a variety of tasks?
>
> While I do consider the _core_ of LaTeX suitable for replacing with
> something more versatile, ConTeXt is quite far behind LaTeX with
> respect to available addons.
Oh-um. AMS-math is the only thing that's lacking and I'm working on
it in my free time (i.e., currently not working on it).
Table creating packages? ConTeXt has the most powerful table
mechanisms I've ever seen; many of them. I can recall off the top of
my head: table, tabular, tabulate, TABLE; overall, they provide for
extrememly complex table typesetting mechanism. Multipage, with
repeating headers and footers, with column spans etc. TABLE is in
almost one-to-one correspondence to the HTML table mechanism.
Multiple footnotes: what exactly do you mean here? I know the ConTeXt
footnote mechanism is in the reworks, being generalized to include an
up-to-now experimental feature (line notes). For what I can see in
the code, the new mechanism allows for multiple (foot)note classes.
Is this it?
As for the thousands of packages: it seems to me that something like
75% of them, if not more, implement features which ConTeXt already
has in the core. I really can't think of anything but math where
ConTeXt lags behind.
David Kastrup
> David Kastrup wrote:
> > Giuseppe Bilotta <bilo...@hotpop.com> writes:
> >
> > > It looks like slowly but surely LaTeX3 is heading towards where
> > > ConTeXt has already been for a long time ;) (Ok, ConTeXt never had
> > > to deal with centuries-old legacies ...)
> >
> > Like amsmath? Like all of the various table creating packages?
> > Like multiple footnote support (to which I'll be adding my own
> > contribution soon)? Like literally thousands of packages
> > available for a variety of tasks?
> >
> > While I do consider the _core_ of LaTeX suitable for replacing
> > with something more versatile, ConTeXt is quite far behind LaTeX
> > with respect to available addons.
>
> Oh-um. AMS-math is the only thing that's lacking and I'm working on
> it in my free time (i.e., currently not working on it).
It's the only thing and you address the other things below? A bit
comical.
> Multiple footnotes: what exactly do you mean here? I know the
> ConTeXt footnote mechanism is in the reworks, being generalized to
> include an up-to-now experimental feature (line notes). For what I
> can see in the code, the new mechanism allows for multiple
> (foot)note classes. Is this it?
With the current state of the art, there are multiple footnote
apparatus possible, in-paragraph footnotes and a few other things
(check the ednotes package and the ledmac package). I'll be adding a
backtracking footnote block breaking algorithm to it that does not
rely on TeX's deficient insertion splitting. It will also enable
working color handling in footnotes split across pages, verbatim use
in the argument of footnote macros, nested footnotes (footnotes within
footnotes), and a few other things. No, it's not a core part of
LaTeX, but an addon. But the addon market for LaTeX is large, with a
few killer apps (bigfoot.sty should become one in the humanities,
listings.sty is for several applications, there are various trickery
stuff alone in the tools directory, journals have their own LaTeX
classes and so on and so on).
> As for the thousands of packages: it seems to me that something like
> 75% of them, if not more, implement features which ConTeXt already
> has in the core. I really can't think of anything but math where
> ConTeXt lags behind.
Well, other people can. And cramming everything in the core does not
make people feel at home with it. One point of the many add-ons of
LaTeX is that they are available when you need them, and you need not
bother with them otherwise.
Jellby
>> As a side-question, is it possible to specify a "constant" line spacing
>> in text? I mean a line spacing that's not affected by an inline formula
>> with too large height/depth. I'd rather have a subscript partly
>> overlapping an ascender than a larger spacing between two lines of text.
>> Or, at least, have the same larger spacing for the whole paragraph.
>
> as a side-question, why would you want it? typesetting to a grid was
> forced upon the typographers of old because of the equipment they used,
> and it looks nice; but where is the evidence that forced spacing
> increases legibility or readability?
Sometimes you have an in-text formula that's a bit too high and it forces a
wider spacing between its line and the previous one. The block of text is
then broken and I find the appearance annoying and... well, "cheap". If the
extra height is not too much, making the line spacing constant could
originate only a small overlap (maybe even unnoticeable) between both
lines, which, in my opinion, wouldn't be so ugly. In some cases this
could/would be preferable.
I'm not saying this is the correct way to fix it, it would be better to
rewrite the paragraph so that the offending formula sits at the first line,
or to write the formula in other way, so that it's not so high. I just
wanted to know whether there is an *option*, a way to accomplish it (other
than enclosing each formula in a zero-height box).
Giuseppe Bilotta
> > > Like amsmath? Like all of the various table creating packages?
> > > Like multiple footnote support (to which I'll be adding my own
> > > contribution soon)? Like literally thousands of packages
> > > available for a variety of tasks?
> > >
> > > While I do consider the _core_ of LaTeX suitable for replacing
> > > with something more versatile, ConTeXt is quite far behind LaTeX
> > > with respect to available addons.
> >
> > Oh-um. AMS-math is the only thing that's lacking and I'm working on
> > it in my free time (i.e., currently not working on it).
>
> It's the only thing and you address the other things below? A bit
> comical.
I address them in case you didn't know the status of the ConTeXt
capabilities.
> > Multiple footnotes: what exactly do you mean here? I know the
> > ConTeXt footnote mechanism is in the reworks, being generalized to
> > include an up-to-now experimental feature (line notes). For what I
> > can see in the code, the new mechanism allows for multiple
> > (foot)note classes. Is this it?
>
> With the current state of the art, there are multiple footnote
> apparatus possible, in-paragraph footnotes and a few other things
> (check the ednotes package and the ledmac package). I'll be adding a
> backtracking footnote block breaking algorithm to it that does not
> rely on TeX's deficient insertion splitting. It will also enable
> working color handling in footnotes split across pages, verbatim use
> in the argument of footnote macros, nested footnotes (footnotes within
> footnotes), and a few other things. No, it's not a core part of
> LaTeX, but an addon. But the addon market for LaTeX is large, with a
> few killer apps (bigfoot.sty should become one in the humanities,
> listings.sty is for several applications, there are various trickery
> stuff alone in the tools directory, journals have their own LaTeX
> classes and so on and so on).
Again, many of those features are present in ConTeXt already.
> > As for the thousands of packages: it seems to me that something like
> > 75% of them, if not more, implement features which ConTeXt already
> > has in the core. I really can't think of anything but math where
> > ConTeXt lags behind.
>
> Well, other people can.
Actually, AFAICS, most people don't even know that ConTeXt exist,
which is why they use LaTeX. Others need advanced math stuff, and for
those AMS-LaTeX is still better.
> And cramming everything in the core does not
> make people feel at home with it. One point of the many add-ons of
> LaTeX is that they are available when you need them, and you need not
> bother with them otherwise.
Oh, I never said that the monolithic approach of ConTeXt was better
than the modular approach in LaTeX. Of course, there is a benefit in
the monolithing approach, in this case, which is that of the various
parts being tightly integrated and consistent, which is not always
the case for the LaTeX modular approach (an example for all: Nath vs
AMS-LaTeX).
Giovanni Zezza
>While I do consider the _core_ of LaTeX suitable for replacing with
>something more versatile, ConTeXt is quite far behind LaTeX with
>respect to available addons.
Well, if you plan to "boldly go where no-one has gone before", chances are
that you will find exactly them there: no-one. No-one of the already
available ones, at least.
The point is: do we need so much a better LaTeX than LaTeX, that we must
face the journey, or we have so many addons around us that we can't move
anymore?
As for "soft moves", or teletransport, I'm inclined to think that claims
about them have been greatly exaggerated.
Ciao.
Donald Arseneau
> extra height is not too much, making the line spacing constant could
> originate only a small overlap (maybe even unnoticeable) between both
> lines, which, in my opinion, wouldn't be so ugly. In some cases this
> could/would be preferable.
\setlength{\lineskiplimit}{ -not_too_much}
Whatever you mean by "not too much"
Donald Arseneau as...@triumf.ca
Dr. Joel M. Hoffman
on a page are fixed to a 12pt grid, my latest attempt to use
"\everypar" to make sure that each paragraph starts at a multiple of
12pt has hit a snag:
LaTeX seems to keep redefining \everypar, so I can't use it. Can
someone help me understand this?
Thanks.
-Joel
Dr. Joel M. Hoffman
And LaTeX also keeps redefining \baselineskip.
Clearly I should have used plain TeX for this, but I didn't, and I
don't want to start over. My first little experiment, just to see how
things work, included:
\raggedbottom
\normalbaselineskip15pt
\baselineskip15pt
\everypar{TEST}
Of these, all but "\raggedbottom" was ignored in my document, no
matter where I put the definitions. Argh!
-Joel
Jellby
>> extra height is not too much, making the line spacing constant could
>> originate only a small overlap (maybe even unnoticeable) between both
>> lines, which, in my opinion, wouldn't be so ugly. In some cases this
>> could/would be preferable.
>
> \setlength{\lineskiplimit}{ -not_too_much}
>
> Whatever you mean by "not too much"
Now, *that's* useful, thank you very much :)
Robin Fairbairns
>>on a page are fixed to a 12pt grid, my latest attempt to use
>>"\everypar" to make sure that each paragraph starts at a multiple of
>>12pt has hit a snag:
>>
>>LaTeX seems to keep redefining \everypar, so I can't use it. Can
>>someone help me understand this?
it uses it to regain control after \section, and similar sorts of
places. that's why paragraph numbering is so damnably difficult.
>And LaTeX also keeps redefining \baselineskip.
yes. every time you select a font.
>Clearly I should have used plain TeX for this, but I didn't, and I
>don't want to start over. My first little experiment, just to see how
>things work, included:
>
> \raggedbottom
> \normalbaselineskip15pt
also reset whenever you select a font.
> \baselineskip15pt
> \everypar{TEST}
>
>Of these, all but "\raggedbottom" was ignored in my document, no
>matter where I put the definitions. Argh!
it's not trivial, but it _is_ doable. take a look at frank's xor
stuff. or switch to context, which (as eny fule kno) has already done
everything, with the footling exception of maths.
Michele Dondi
>> As for the thousands of packages: it seems to me that something like
>> 75% of them, if not more, implement features which ConTeXt already
>> has in the core. I really can't think of anything but math where
>> ConTeXt lags behind.
>
>Well, other people can. And cramming everything in the core does not
>make people feel at home with it. One point of the many add-ons of
>LaTeX is that they are available when you need them, and you need not
>bother with them otherwise.
I have never tried ConTeXt, and I'm perfectly fine with LaTeX, but I'd
feel at home with more stuff *consistently* moved (i.e. not "crammed")
into the core. The good point of LaTeX packages that you mention is
not in contraddiction with the fact that many of them may well fit in
the LaTeX distribution itself rather than being "external" as
currently are: people would continue to load them only when needed.
As a side note, I don't know if technically this is feasible and if so
how easily, but one could envision a transparent modular system of
packages. For example we may have enhanced classes that allow for
headings tuning: all related commands are contained in a (sort of)
"package". But the "package" is loaded automatically on the first
appearance of one of those commands.
The point is that there are so many modules that most users just
beyond the newbie level use more or less by default... it seems that
classes like memoir or KOMA* not only provide alternative styles and
appearences, but are oriented exactly at offering all those features
all-in-one, so to say. And more importantly, in a *consistent* manner,
for what is possible.
Last, to make an analogy, the current situation with LaTeX reminds me
of the *nix cmd line utilities: many, useful, flexible, modular,
fantastic! But also having duplicate functionalities (in some cases),
inconsitent systems of switches, limitations, etc.
As I learned Perl I had the impression that all those powerful utils
had been gathered in one tool in a *consistent* manner
(notwithstanding what you may have heard about Perl!) to a synergetic
effect: it is possible to do millions of things along a pipe, but in
Perl you can connect your pipes in just as complex a network as you
like...
What I wrote above doesn't mean that the utils become suddenly
obsolete: I still use them stand-alone or even in things like
perl -lne 'next unless s/\d+://; print for split /,/' | wc -l
simply because taking care of a counter myself would be more expensive
in terms of keystrokes, and I am lazy! (so I have at least one of the
three principal virtues of a programmer!)
So I don't say a more-stuff-in-the-core LaTeX shouldn't support
further "external" extentions. Only that there are features that most
people consider to be basic for their document design: most notably
headings and sectioning commands customization, "geometry" fine
tuning, etc. It wouldn't be so bad to have such things in LaTeX by
default.
Michele
--
>It's because the universe was programmed in C++.
No, no, it was programmed in Forth. See Genesis 1:12:
"And the earth brought Forth ..."
- Robert Israel on sci.math, thread "Why numbers?"
William F. Adams
pass a LaTeX-tagged source file into ConTeXt and directly typeset from that.
Glad others have mentioned the XO float experimental code and LaTeX3 nascency.
In addition to being a matter of (traditional) aesthetics, there's also the
practical consideration of minimizing the distraction of any showthrough from
backing pages and the more important one of such being a publisher's style, or
part of a designer's specification.
IME, limiting glue to above and below floats works as a good start and gets one
a fair bit of the way there, as will adding space at need above heads. A
judicious ``\enlargethispage{+-\baselineskip}'' (in pairs) will get one even
farther along, then one merely need fix any pages where an equation falls at
the bottom of a page (TeX's idea of where the baseline is in an equation
doesn't match mine) will finish things up.
William
--
William Adams
http://members.aol.com/willadams
Sphinx of black quartz, judge my vow.