# Bottom margin vs. descenders question

Philip Taylor P.Taylor at Hellenic-Institute.Uk
Thu Jun 25 17:43:54 CEST 2020

Barbara Beeton wrote:

> An even more rigid model followed by traditional typesetting is "grid
> typesetting", in which *all* baselines line up on facing pages. As has
> often been noted, this is *not* followed by, and is not easy to obtain
> with TeX, especially in the presence of math displays.

As Barbara is (in my experience) never wrong, what follows is in no
sense a correction but rather an amplification.  One /can/ typeset
spreads on a grid, but TeX lacks (IMHO) one vital element that would
much simplify this process, to wit a box with two reference points.  In
the absence thereof one has to work around the problem (an early
solution made use of \pseudoboxes) but I think that were I to tackle the
problem again today I would take advantage of the fact that leaders
/behave/ like a skip but can actually include real typeset material.

To re-cycle my example of earlier this afternoon, consider the following :

> % !TeX Program=XeTeX
>
>
> \parskip = 0 pt
>
> \parindent = 0 pt
>
>
> \font \gfont = cmr10 scaled 5000
>
> \def \good {\lower 3,5 ex \hbox {\gfont g}ood }
>
>
> \baselineskip = 5mm
>
> \lineskip = 0 pt
>
> \lineskiplimit = -\maxdimen
>
> \abovedisplayskip = 0 pt
>
> \abovedisplayshortskip = 0 pt
>
> \belowdisplayskip = 0 pt
>
> \belowdisplayshortskip = 0 pt
>
>
> \topglue 12.2mm
>
> \obeylines
>
> The Leith police dismisseth us, for we sufficeth them.
>
> The quick brown fox jumps right over the lazy dog's back.
>
> \leaders \vbox {$$x^2 = y^2 + z^2$$}\vskip 2\baselineskip
>
> Now is the time for all \good men to come to the aid of the party.
>
> Here endeth the first lesson.
>
> \eject
>
>
> \end
>

View the output in Adobe Acrobat with the grid visible, grid separation
5mm, sub-divisions 10, and you will see that baseline registration is
maintained perfectly across the display maths.

/Philip Taylor///[E&OE — the first time I have tried this technique.]
-------------- next part --------------
An HTML attachment was scrubbed...