Received: by nummer-3.proteosys id <01C19443.3E74C3CC@nummer-3.proteosys>; Thu, 3 Jan 2002 11:41:47 +0100 MIME-Version: 1.0 x-vm-v5-data: ([nil nil nil nil nil nil t t nil][nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil nil]) Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C19443.3E74C3CC" X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message Subject: Standard styles Date: Sun, 3 Mar 1991 20:11:50 +0100 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: From: "LaTeX-L Mailing list" Sender: To: "Rainer M. Schoepf" Reply-To: "LaTeX-L Mailing list" Status: R X-Status: X-Keywords: X-UID: 296 This is a multi-part message in MIME format. ------_=_NextPart_001_01C19443.3E74C3CC Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable At Cork, I think that Frank mentioned the idea of both: 1. supplying style files that emulate the effect of the present LaTeX = 2.09 "standard styles" (for backwards compatibility) 2. supplying new "standard style" files. I think he mentioned having analogues of the current article, report and book, plus having a "conference proceedings" style. This note is about (2). In general, I'd suggest that (2) be done in line with traditional = mainstream publishing practice. [This seemed to be what Phil Taylor was after = when, at Cork, he described the adjustments he'd had to make to get a LaTeX-ed book that didn't scream "I've been produced by (La)TeX!", and could be published. It would be nice if people in his position didn't have to = make so many adjustments.] This leaves the problem of determining what = mainstream practice is. Here are some comments about page-sizes and typefaces. JOURNAL ARTICLES As regards journal articles, the book by Page, Campbell & Meadows may be helpful [1, pp. 35-6]. The gist of it seems to be that, for reasons connected with the sizes of printing presses, journals tend to ignore B5 and A4 (or did when [1] was written) and go for page sizes of: * 244 mm x 172 mm, with a 2-column layout. This is slightly smaller than B5 (because B5 doesn't make optimum use of the presses). Page et al. seem to be talking in terms of typesize of "9 on 10.5" for such a design (although strictly speaking this was for a B5 = example.) Perhaps this is what the analogue of = \documentstyle[9pt,twocolumn]{article} should be designed for. * 276 mm x 219 mm (demi quarto), with a 2-column layout. This is = slighly smaller than A4. Page et al. seem to be talking in terms of typesize = of "10 on 11.5" for such a design. Perhaps this is what the analogue of = \documentstyle[10pt,twocolumn]{article} should be designed for. Page et al. also give two B5 (250 mm x 176 mm) examples: * single-column, type-area 197 mm x 130 mm, 10 on 11.5 Perhaps the analogue of \documentstyle[10pt]{article} should implement something like this. * double-column, type-area 206 mm x 133 mm, 9 on 10.5 These may be more to illustrate economics of different designs than to say that people actually use B5 much, though. They say that single column with 10pt is usually preferred for maths & physics. If some of the "new standard style" files implemented designs aimed at the above sizes of paper (with crop marks to show the corners of the target area), they might serve as a good starting point for anyone = who has to produce a style file for a real journal, particularly if the rest of the design was based on "the average design" of some real journals Such styles might also keep authors happy who want to see what their = paper might look like in a real journal. BOOKS As regards books, Hugh Williamson suggests that the A series haven't = caught on for book work (not in 1983 Britain, anyway) [2, ch. 3]. He lists the British Standard cut-page sizes, including (in millimetres): quarto octavo crown 246 x 189 186 x 123 large crown 258 x 201 198 x 129 demy 276 x 219 216 x 138 royal 312 x 237 234 x 156 Apparently 181 x 111 and 178 x 111 are used for paperbacks. He also lists the corresponding American stock sheet sizes, which give cut-page sizes of: 140 x 216 156 x 235 127 x 187 137 x 203 140 x 210 143 x 213 which agrees with the Chicago Manual of Style [3, p. 623]. Obviously there is a great variety of sizes. Presumably any new = "standard book styles" would have to be designed for a particular size of paper. = I'd suggest that they be designed for a cut-sheet size that IS actually used = for books (e.g. one or more of the above) and that crop-marks be produced to show the "target area". Ruari McLean [4, p. 130] says that demi octavo = is one of the most "normal" sizes for books, so perhaps 216 x 138 (or the nearest US size, 216 x 140) should be one of the "target areas". As regards section headings - Williamson [2, p. 163] mentions the scheme: level A (i.e., \section) in roman capitals; level B in small capitals; level C in upper/lower case italic, level D in upper/lower case italic, followed by a point, run-in with text. "The Chicago Manual of Style" [3, p. 570] mentions: level A caps & small caps (or full caps) level B in small capitals level C in italics, upper/lower case, followed by a point, run in with text. There seems a fair amount of consensus between these gurus. Perhaps elements of these schemes could be combined to give a mainstream design that isn't going to upset anyone. A4 (and US equivalent) We have the contradition that: * LaTeX is a typeSETTING system. For most books and journals, the typesetting tradition uses paper that is smaller than A4, and uses fonts of around 10pt. * most LaTeX output comes, at least in the first instance, on A4 (or US equivalent) paper from a laserprinter, where A4 etc. is an = "office" paper size that fits in with the typeWRITING tradition (which usually involves fonts of arout 12pt). So the naive user, seeing something from a "book" \documentstyle on A4 in 10pt, starts to ask "Why doesn't it use all the paper?". Crop marks, to indicate the page-size for which the design is intended, might help: * such people to understand why the design doesn't fill A4 * avoid such people getting an a4.sty out of an archive and going \documentstyle[a4]{...} (to get text height/width that purports to be "for A4 paper") and then wondering why they've got something that is difficult to read * help anyone who wants to produce a book to visualise the effect that was intended by the designer. Although it may not be clear which, of the large variety of cut-page sizes in common use, should provide the target "cut-page sizes" for the "new standard style's" article and book designs (and their variants), it does seem fairly clear that: * the target "cut-page sizes" should generally be smaller than the A4 = sheet on which output will initially appear * the users should be made aware that the target area is different from = A4, so they don't ruin the design in their attempts to "make it fit A4". Obviously, if a design is intended for A4 (as it might be for a draft article, a report or a thesis), the crop marks would be omitted. Users would then know that they can use the output just as it comes out of their laserprinter. GENERAL I'd be inclined to make Times Roman the normal font for running text in any "new standard styles" (with a switch somewhere to substitute Computer Modern for anyone who doesn't have Times Roman, or wants to get "nearly ready for publication" before they switch from preview-able Computer Modern to less-easily-previewed Times Roman). I have the impression that Times Roman is "the default font" for running text in mainstream publishing. This would cut out one change that = publishers often seem (rightly or wrongly) to ask people to make who are trying to typeset a book themselves. [I have nothing against Computer Modern. I'm just inclined to "bow to the inevitable".] I'd be inclined to refrain from providing style-files in situations where they are unrealistic. For example, if it is very rare for real journals and books to use a 12 point typeface for running text, it may not be worth the effort of supporting = \documentstyle[12pt]{article} and \documentstyle[12pt]{book}. To support such things may involve asking the question "What design would one have for a 12pt journal?", to which the real answer may be "One wouldn't actually have a 12pt = journal." As an alternative to "making the cut-page size depend on the font-size" (which seems the consequence of the 2.09 way of doing things), it might be worth considering "making the font-size depend on the target = page-size". E.g. rather than having the user go \documentstyle[11pt,twocolumn]{article} and (having measured distance between the crop-marks) deducing the size of paper for which the design is intended, it might make more sense for the user to go \documentstyle[...b5]{article} and get a one-column 10 on 11.5 design intended for B5 paper, or \documentstyle[...b5,twocolumn]{article} and get a two-column 9 on 10.5 design for B5, or \documentstyle[...demisemioctavo]{article} to get an error message along the lines "Sorry. No single-column design is available for the demisemioctavo paper-size." Designs could then be limited to combinations of paper-size and layout that a real journal might conceivably use. REFERENCES [1] Gillian Page, Robert Campbell & Jack Meadows. "Journal publishing: principles and practice", Butterworths, 1987. ISBN 0-408-10716-2. [2] Hugh Williamson. "Methods of Book Design", Yale University Press, 1983. ISBN 0-300-03035-5. [3] "The Chicago Manual of Style", Chicago University Press, 1982. ISBN 0-226-10390-0. [4] Ruari McLean, "The Thames and Hudson Manual of Typography", Thames and Hudson, 1980. ISBN 0-500-68022-1. ------------ David Rhead (JANET: d.rhead@uk.ac.nottingham.ccc.vme) ------_=_NextPart_001_01C19443.3E74C3CC Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Standard styles

At Cork, I think that Frank mentioned the idea of = both:
1.  supplying style files that emulate the = effect of the present LaTeX 2.09
    "standard styles" (for = backwards compatibility)
2.  supplying new "standard style" = files.  I think he mentioned having
    analogues of the current article, = report and book, plus having
    a "conference = proceedings" style.
This note is about (2).

In general, I'd suggest that (2) be done in line with = traditional mainstream
publishing practice.  [This seemed to be what = Phil Taylor was after when,
at Cork, he described the adjustments he'd had to = make to get a LaTeX-ed
book that didn't scream "I've been produced by = (La)TeX!", and could be
published.  It would be nice if people in his = position didn't have to make so
many adjustments.]  This leaves the problem of = determining what mainstream
practice is.  Here are some comments about = page-sizes and typefaces.


          &nbs= p;            = ;    JOURNAL ARTICLES

As regards journal articles, the book by Page, = Campbell & Meadows may be
helpful [1, pp. 35-6].  The gist of it seems to = be that, for reasons
connected with the sizes of printing presses, = journals tend to ignore
B5 and A4 (or did when [1] was written) and go for = page sizes of:
*  244 mm x 172 mm, with a 2-column = layout.  This is slightly smaller
   than B5 (because B5 doesn't make optimum = use of the presses).
   Page et al. seem to be talking in terms = of typesize of "9 on 10.5"
   for such a design (although strictly = speaking this was for a B5 example.)
   Perhaps this is what the analogue of = \documentstyle[9pt,twocolumn]{article}
   should be designed for.
*  276 mm x 219 mm (demi quarto), with a = 2-column layout.  This is slighly
   smaller than A4.  Page et al. seem = to be talking in terms of typesize of
   "10 on 11.5" for such a = design.
   Perhaps this is what the analogue of = \documentstyle[10pt,twocolumn]{article}
   should be designed for.

Page et al. also give two B5 (250 mm x 176 mm) = examples:
*  single-column, type-area 197 mm x 130 mm, 10 = on 11.5
   Perhaps the analogue of = \documentstyle[10pt]{article} should
   implement something like this.
*  double-column, type-area 206 mm x 133 mm, 9 = on 10.5
These may be more to illustrate economics of = different designs than
to say that people actually use B5 much, = though.  They say that
single column with 10pt is usually preferred for = maths & physics.

If some of the "new standard style" files = implemented designs aimed
at the above sizes of paper (with crop marks to show = the corners of
the target area), they might serve as a good starting = point for anyone who
has to produce a style file for a real journal, = particularly if the rest
of the design was based on "the average = design" of some real journals
Such styles might also keep authors happy who want to = see what their paper
might look like in a real journal.


          &nbs= p;            = ;           = BOOKS

As regards books, Hugh Williamson suggests that the A = series haven't caught
on for book work (not in 1983 Britain, anyway) [2, = ch. 3].  He lists the
British Standard cut-page sizes, including (in = millimetres):
          &nbs= p;            = ;       = quarto            = octavo
crown          = ;            =   246 x 189        186 x = 123
large = crown           &n= bsp;      258 x = 201        198 x 129
demy          =             &= nbsp;  276 x 219        216 x = 138
royal          = ;            =   312 x 237        234 x = 156
Apparently 181 x 111 and 178 x 111 are used for = paperbacks.
He also lists the corresponding American stock sheet = sizes, which give
cut-page sizes of:
      140 x 216
      156 x 235

      127 x 187
      137 x 203
      140 x 210
      143 x 213
which agrees with the Chicago Manual of Style [3, p. = 623].

Obviously there is a great variety of sizes.  = Presumably any new "standard
book styles" would have to be designed for a = particular size of paper.  I'd
suggest that they be designed for a cut-sheet size = that IS actually used for
books (e.g. one or more of the above) and that = crop-marks be produced to
show the "target area".  Ruari McLean = [4, p. 130] says that demi octavo is
one of the most "normal" sizes for books, = so perhaps 216 x 138 (or the
nearest US size, 216 x 140) should be one of the = "target areas".

As regards section headings -
Williamson [2, p. 163] mentions the scheme:
      level A (i.e., = \section) in roman capitals;
      level B in small = capitals;
      level C in upper/lower = case italic,
      level D in upper/lower = case italic, followed by a point, run-in
          &nbs= p; with text.
"The Chicago Manual of Style" [3, p. 570] = mentions:
      level A caps & = small caps (or full caps)
      level B in small = capitals
      level C in italics, = upper/lower case, followed by a point,
          &nbs= p; run in with text.
There seems a fair amount of consensus between these = gurus.  Perhaps
elements of these schemes could be combined to give a = mainstream design
that isn't going to upset anyone.


          &nbs= p;            = ; A4 (and US equivalent)

We have the contradition that:
*  LaTeX is a typeSETTING system.  For most = books and journals,
   the typesetting tradition uses paper = that is smaller than A4, and
   uses fonts of around 10pt.
*  most LaTeX output comes, at least in the = first instance, on A4 (or
   US equivalent) paper from a = laserprinter, where A4 etc. is an "office"
   paper size that fits in with the = typeWRITING tradition (which usually
   involves fonts of arout 12pt).
So the naive user, seeing something from a = "book" \documentstyle on A4
in 10pt, starts to ask "Why doesn't it use all = the paper?".  Crop
marks, to indicate the page-size for which the design = is intended,
might help:
*  such people to understand why the design = doesn't fill A4
*  avoid such people getting an a4.sty out of an = archive and going
   \documentstyle[a4]{...} (to get text = height/width that purports
   to be "for A4 paper") and then = wondering why they've got something
   that is difficult to read
*  help anyone who wants to produce a book to = visualise the effect
   that was intended by the = designer.

Although it may not be clear which, of the large = variety of cut-page
sizes in common use, should provide the target = "cut-page sizes" for the
"new standard style's" article and book = designs (and their variants),
it does seem fairly clear that:
*  the target "cut-page sizes" should = generally be smaller than the A4 sheet
   on which output will initially = appear
*  the users should be made aware that the = target area is different from A4,
   so they don't ruin the design in their = attempts to "make it fit A4".

Obviously, if a design is intended for A4 (as it might = be for a
draft article, a report or a thesis), the crop marks = would be omitted.
Users would then know that they can use the output = just as it comes out
of their laserprinter.


          &nbs= p;            = ;       GENERAL

I'd be inclined to make Times Roman the normal font = for running text
in any "new standard styles" (with a switch = somewhere to substitute
Computer Modern for anyone who doesn't have Times = Roman, or wants
to get "nearly ready for publication" = before they switch from
preview-able Computer Modern to less-easily-previewed = Times Roman).
I have the impression that Times Roman is "the = default font" for running
text in mainstream publishing.  This would cut = out one change that publishers
often seem (rightly or wrongly) to ask people to make = who are trying to
typeset a book themselves.  [I have nothing = against Computer Modern.
I'm just inclined to "bow to the = inevitable".]

I'd be inclined to refrain from providing style-files = in situations
where they are unrealistic.  For example, if it = is very rare for
real journals and books to use a 12 point typeface = for running text,
it may not be worth the effort of supporting = \documentstyle[12pt]{article}
and \documentstyle[12pt]{book}.  To support such = things may involve
asking the question "What design would one have = for a 12pt journal?", to
which the real answer may be "One wouldn't = actually have a 12pt journal."

As an alternative to "making the cut-page size = depend on the font-size"
(which seems the consequence of the 2.09 way of doing = things), it might
be worth considering "making the font-size = depend on the target page-size".
E.g. rather than having the user go
      = \documentstyle[11pt,twocolumn]{article}
and (having measured distance between the crop-marks) = deducing the
size of paper for which the design is intended, it = might make more
sense for the user to go
      = \documentstyle[...b5]{article}
and get a one-column 10 on 11.5 design intended for = B5 paper, or
      = \documentstyle[...b5,twocolumn]{article}
and get a two-column 9 on 10.5 design for B5, = or
      = \documentstyle[...demisemioctavo]{article}
to get an error message along the lines = "Sorry.  No single-column
design is available for the demisemioctavo = paper-size."  Designs could
then be limited to combinations of paper-size and = layout that a real
journal might conceivably use.


          &nbs= p;            = ;     REFERENCES

[1]   Gillian Page, Robert Campbell & = Jack Meadows.
      "Journal = publishing: principles and practice", Butterworths, 1987.
      ISBN = 0-408-10716-2.

[2]   Hugh Williamson.  "Methods = of Book Design", Yale University Press,
      1983. ISBN = 0-300-03035-5.

[3]   "The Chicago Manual of = Style", Chicago University Press, 1982.
      ISBN = 0-226-10390-0.

[4]   Ruari McLean, "The Thames and = Hudson Manual of Typography",
      Thames and Hudson, = 1980.  ISBN 0-500-68022-1.

          &nbs= p;            = ;       ------------

David Rhead
(JANET: d.rhead@uk.ac.nottingham.ccc.vme)


------_=_NextPart_001_01C19443.3E74C3CC--