MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
References: <18249.57441.391405.468407@morse.mittelbach-online.de>
            <008a01c83026$f01a3900$14b2a8c0@DOMINUS>
            <18250.53690.194331.660833@morse.mittelbach-online.de>
            <005401c8305d$cf23a290$14b2a8c0@DOMINUS>
            <18251.16134.90835.686471@morse.mittelbach-online.de>
            <006a01c830c9$d1352560$14b2a8c0@DOMINUS>
            <18251.53934.527003.575084@morse.mittelbach-online.de>
            <008401c83127$c6f994b0$14b2a8c0@DOMINUS>
Message-ID:  <18252.29644.250197.671360@morse.mittelbach-online.de>
Date:         Tue, 27 Nov 2007 20:45:16 +0100
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
Sender: Mailing list for the LaTeX3 project <LATEX-L@LISTSERV.UNI-HEIDELBERG.DE>
From: Frank Mittelbach <frank.mittelbach@LATEX-PROJECT.ORG>
Subject: Re: Extending the output routine to middle floats
To: LATEX-L@LISTSERV.UNI-HEIDELBERG.DE
In-Reply-To:  <008401c83127$c6f994b0$14b2a8c0@DOMINUS>
Precedence: list
Status: R

Ulrich Dirr writes:
 > >  > this should be covered by my above definition (text above
 > >  > 16\baselineskip, float area 6\baselineskip, text after  
 > >  > 42-(16+6)\baselineskip). 
 > > 
 > > as one possibility for specification (not the actual values but the
 > > concept: which is fixed starting position with a given vertical size and
 > > one area only). my question is what others should/could be supported?
 > 
 > not one area but areas which can be in 0 to n columns with and without
 > spans. 

yes sure I should have been more explicit here. I meant more than one area
vertically in a single column (or as spans).

allowing for several middle columns that use or span different columns is
something i though was obvious, eg. just like the current algorithm supports
several top areas as long as they don't overlap --- the middle part would
support several middle columns as long as they don't overlap (with or without
spans).

in theory one could probably also make overlapping areas possible (in terms of
the columns they use) but then rules for deciding how they are placed would
become very difficult to describe).

 > >  > I thought this was what I'd proposed.
 > > 
 > > yes, I think so, you proposed option 2. But what about:
 > > 
 > >  > >     * ...other ideas...
 > > 
 > > is there any kind of reasonable rule set that is not covered
 > > by my initial variations and should perhaps be supported as well?  
 > 
 > I don't think so. O.k., maybe we can think of an explicity given
 > vertical offset which is added or subtracted after each column, e.g.

not that i think this would be very attractive other than for some weird
magazine layout :-) but in any case it would automatically supported already
since you could specify different vertical offsets for each column area
individually.

 > a magazine designer might like this idea to have picture material going
 > like a sinus wave from page to page ... ;-) Or picture material always
 > going in the outside column and going down somewhat after each chapter
 > (like an index thumb).

same situation, would work well enough if you really want to, since the new
design would allow you to change your float setup (ie which areas to use where
and when) so that you could implement something like this changing the
positioning after each chapter

 > I will zap through some design books -- maybe I will find some other
 > interesting variations.


ok, thanks, but it is good to know that essentially so far nobody came up with
something radically different and missing. so only remains to make it work
(right now it accepts the area but doesn't then place the float :-)

frank