Received: by nummer-3.proteosys id <01C19443.465783C4@nummer-3.proteosys>; Thu, 3 Jan 2002 11:42:00 +0100 MIME-Version: 1.0 x-vm-v5-data: ([nil nil nil nil nil nil nil 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_01C19443.465783C4" X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message Subject: Should tables expand to width of text-column? Date: Thu, 5 Sep 1991 01:00:00 +0100 Message-ID: X-MS-Has-Attach: X-MS-TNEF-Correlator: From: Sender: "LaTeX-L Mailing list" To: "Rainer M. Schoepf" Reply-To: "LaTeX-L Mailing list" Status: R X-Status: X-Keywords: X-UID: 379 This is a multi-part message in MIME format. ------_=_NextPart_001_01C19443.465783C4 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Chris Hamlin asked for facilities to fit a table/tabular to the maximum width available, since this is the style for Physical Review tables. I've found a design-guru who gives a plausible argument against styles = such as that of Physical Reviews. In "Graphic Design for the Electronic = Age", Jan White says (page 143) ... good sense not to set them the same width as the column of = type ... welcome interruption in the stream of text ... more compelling reason ... legibility ... tables should = never be arbitrarily spaced out to fit the overall width. Artificial = gaps impede reading. Tables should be set to their natural width. I find this fairly convincing. It seems analogous to the convention in ordinary text that "the inter-word space should never be more than the inter-line space". This is a situation where I wouldn't worry too much about the argument = that "troff+tbl can do it, so LaTeX should be able to do it". Maybe the = authors of tbl made a mistake in providing support for a feature of bad design? If LaTeX 3.0 did provide more support for this sort of thing than LaTeX 2.09's tabular* and \extracolsep, I'd be inclined to keep it low-profile (e.g., documented somewhere obscure) so that people whose house-style = says that tables must "expand to text-width" have the facilities, but people = who have the choice normally set tables to their "natural width". David Rhead JANET: d.rhead@uk.ac.nottingham.ccc.vme ------_=_NextPart_001_01C19443.465783C4 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Should tables expand to width of text-column?

Chris Hamlin asked for facilities to fit a = table/tabular to the maximum
width available, since this is the style for Physical = Review tables.

I've found a design-guru who gives a plausible = argument against styles such
as that of Physical Reviews.  In "Graphic = Design for the Electronic Age",
Jan White says (page 143)
     ...  good sense not to = set them the same width as the column of type
     ...  welcome = interruption in the stream of text

     ...  more compelling = reason ...  legibility ...  tables should never
     be arbitrarily spaced out to = fit the overall width.  Artificial gaps
     impede reading.  Tables = should be set to their natural width.
I find this fairly convincing.  It seems = analogous to the convention in
ordinary text that "the inter-word space should = never be more than the
inter-line space".

This is a situation where I wouldn't worry too much = about the argument that
"troff+tbl can do it, so LaTeX should be able to = do it".  Maybe the authors
of tbl made a mistake in providing support for a = feature of bad design?

If LaTeX 3.0 did provide more support for this sort of = thing than LaTeX
2.09's tabular* and \extracolsep, I'd be inclined to = keep it low-profile
(e.g., documented somewhere obscure) so that people = whose house-style says
that tables must "expand to text-width" = have the facilities, but people who
have the choice normally set tables to their = "natural width".


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


------_=_NextPart_001_01C19443.465783C4--