[texhax] could/should I have used TeX/LaTeX? (long post)

[James Miller]

I'm on the verge of completing my dissertation, which has been written 
using OpenOffice.org. I ran into many problems and program shortcomings, 
and had to learn alot about desktop publishing to use it. I still consider 
my knowledge about these things spotty. I'd like to ask in this post if I 
might have been able to use TeX/LaTeX for my project.

Let me start off by saying that I was looking at Tex/LaTeX toward the 
beginning (4 years ago or so) but found the documentation over my head. 
Now that I've been using Linux for about 5 years and have learned 
associated computing concepts better, I find the documentation more 
comprehensible. I would really have liked to use TeX/LaTeX and am 
wondering now if I should have another go at it: I like the idea of 
producing marked-up text, and am sort of a hands-on type guy. But to be 
truthful, I'm just not sure anything but a WYSIWYG word processor could do 
what I needed. At the least, I would have had to do some things pretty
differently.

To give an idea of my project, I'll offer a verbal description as well as 
providing some links to screenshots of particularly complex parts of my 
document (NOTE: the screenshot images are rather large--dialup users might 
want to skip them). This was a dissertation in the humanities, and one 
that required fairly extensive use of foreign language fonts. The thing 
that turned me away from TeX/LaTeX in the beginning was my need to use a 
particular type of Greek font. My style sheet dictates that I use fonts 
provided by the SBL (Society of Biblical Literature). These are truetype 
fonts. And I needed a special type of truetype Greek font to boot--one 
that bears a similarity to what is know as "uncial" script. This is 
different from Greek capitals in several respects, by the way, so Greek 
capitals were not an option.

The first segment of my dissertation would not have posed particular 
problems, I think. There, I give a narrative description of the problem 
and how I intend to address it. Occasional Greek (both the uncial as well 
as a more regular polytonic Greek font with diacriticals) text appears 
there, as well as a bit of Hebrew (no vowel pointing). Some French and 
German is cited on occasion and there are various tables, some of which 
have images inserted in them. The first part of the dissertation seems 
like it would pose less challenges trying to learn and utilize TeX/LaTeX.

The second part, however, utilizes some unusual page layouts. In addition, 
a large portion of this second part (in terms of megabytes on disk) is 
comprised of images inserted into tables--about 75 pages of this. The 
images are photographic reproductions of a manuscript I worked with. Other 
parts of this lengthy second section are landscape-oriented pages with 
either a table of four columns and 100 rows, or two pseudo-columns 
containing Greek text in two different font faces.

A screenshot of the part with the table of four columns and 100 rows can 
be seen here: http://cmiller.sdf-us.org/Exh-A.png . I had to landscape 
orient the pages so all the data I wanted to present in the table would 
fit on a single horizontal line. The layout there is not as complex as in 
the next 2 sections, so trying to do this sort of table with TeX/LaTeX 
might not be so much of a challenge.

A screenshot of the next section--the one containing alot of Greek text in 
two parallel pseudo-columns, one of which uses the uncial font I described 
earlier--can be seen at http://cmiller.sdf-us.org/Exh-B.png . I refer to 
the two columns of Greek text as "pseudo-columns" because they are not 
really columns: the column starting on the right half of that page is not 
a continuation of the column on the left, but a continuation of the column 
on the right hand side of the preceding page. Likewise with the column on 
the left: it picks up from where the column on the left hand side of the 
preceding page left off. This is what could be referred to as a "parallel 
text," though that usually refers to a text with translation (like in the 
Loeb series) while this is simply the same text rendered in different 
fonts and with some minor formatting changes. This section should have 
been done as one long table of one row and two columns. It could not be 
done like this however, owing to OpenOffice's inability to make table 
columns span pages. I had to apply a hack using various tabs and macros to 
get this sort of parallel appearance: I didn't like having to do that, but 
since I'd already cast in my lot with OpenOffice, I carried through with 
it. Note that there is even a sort of third pseudo-column at the right 
margin with some numerals bracketed by arrows.

A screenshot of the third part of this second section, the one with all 
the images of the manuscript, can be seen here: 
http://cmiller.sdf-us.org/Appfull-17.png . This was probably the most 
problematic part of the dissertation: I had to essentially recreate it 3 
times owing to problems with OpenOffice. The problem has to do with 
positioning of the images on the page. The images vary somewhat in size, 
and I wanted them centered to page. That required inserting them in what 
OpenOffice calls "frames" and, furthermore, anchoring those frames to the 
page. The problem with that approach is that when you anchor to page like 
that, you can no longer insert pages ahead of those with page-anchored 
frames. If you try to insert a document with page-anchored frames into 
another, all kinds of formatting weirdness ensues. I'll spare you the rest 
of the gory details. I've anchored them to paragraph for the time being, 
and am contemplating a switchover to OpenOffice 2 where this 
page-centering can be done more sanely. Note that the image has numbers 
running down its right margin: these are verse numbers and their alignment 
with parts of the text is crucial within certain limits. To do this, I 
created a table whose left column was occupied by the image made as wide 
as the page margins would allow while still leaving room for the numbering 
going down the right. The numbering down the right side is actually the 
contents of the right column of the table.

I'm going to stop my summary there, since it's probably already become 
tedious enough. I want to ask in closing whether I might have expected to 
use TeX/LaTeX to create this document. Would there have been advantages 
over using it as opposed to the WYSIWYG program I chose? Specifically, how 
would I have dealt with the uncial font issue? Would I not have needed to 
convert that truetype font to something TeX/LaTeX could use first? What 
kind of advantages might I have had in creating the second part of my 
document with the weird page layouts using TeX/LaTeX? For the part with 
all the images I'm guessing that, to get the right placement of the 
numerals down the right side I would have needed to take a quite different 
approach: I would probably have needed to insert those numerals right into 
the jpgs using Gimp or its lousy commercial counterpart :), then insert 
the whole image, right? I can't imagine how the right alignment could be 
accomplished without some form of WYSIWYG tool. And how about getting the 
right alignment for the text in the pseudo-columns of part 2 of the second 
section? I suppose that would be done by introducing & symbols for the 
column divider, then stipulating a line end with \\ at the end--that is, 
if the font issue could be surmounted. Since the presentation in this 
second part is highly visual, perhaps it demands an overall WYSIWYG 
approach?

I may end up writing other large documents like this and I'm still 
attracted by the mark-up approach and feel I have a much better 
understanding of it now. Input on how/whether I might have done those 
things I ended up doing with the WYSIWYG program will be appreciated and 
will help me decide whether to invest any more time and effort on 
TeX/LaTeX. I keep getting the impression that it's for natural science 
people or technical/engineering types more than for humanities scholars 
like myself, so input from other humanities users would be especially 
valuable.

Thanks in advance for any input. My apologies if this post 
detracts/distracts from the purpose of this list.

James