[OS X TeX] New alpha Textures with support for graphics

[Bruno Voisin]

Le 13 juil. 06 à 19:32, Claus Gerhardt a écrit :

> Before you get overboard with your praise of Textures, [...]

OK, here's my take on Textures.


Pros
====

- Being self-contained. My first contact with TeX was in 1991 when I  
was handed the Textures box, containing: the Textures floppies, the  
CM/PS font floppies, the Textures User's Guide, the TeXbook, the  
Textures LaTeX Local Guide, the LaTeX manual. These were all the  
tools and documentation that were necessary for using (La)TeX to  
produce (beautiful) mathematical documents.

With this box, I have been able for say 3 years to write all the  
mathematical prose I had to write. In retrospect these were the most  
productive years of my professional life. That without using the  
internet for more than sending the odd email once a fortnight, and  
without having any contact with another TeXist or access to any other  
documentation. And possibly because of this.

Sure, I was missing many of the more advanced functionalities that  
TeX had to offer, related for example to dvips or to all the LaTeX  
styles developed around the world. But on the other hand it was  
possible to be perfectly functional without these. There's always a  
balance to find between the time spent to explore advanced  
functionalities, and the need one really has for them.

That, sadly, ceased to be possible when I started to collaborate with  
other people (and edit the proceedings of a conference) in 1994. See  
the Cons below.

[Aside: Similarly, there is a trend in the open-source world to  
consider that when one plans to use a software, one should join the  
corresponding mailing list and consider this list as part of the  
documentation. I don't agree with this. A software distro should be  
self-contained, and contain in particular all the documentation and  
components necessary for the software to be fully functional. A  
mailing list is a very useful thing (for example this list), allowing  
to exchange with other users and to learn about the most advanced  
functionalities of and extensions to the software; but it should be  
an added bonus, not a requirement for using the software. Again,  
there is a balance to find between the time spent to listening and  
posting to a list, and the amount of useful info one gets from it. I,  
for one, fail spectacularly at finding this balance.]

- Brand loyalty. AFAIK Textures invited the modern GUI for TeX: a  
window for .tex input, a window for .dvi output, a window for .log  
console, save and print from the File menu, edit from the Edit menu,  
compile from the Typeset menu, view from the View menu. All this in a  
clean and elegant interface, integrated within one application. With  
Textures, TeX appeared just like any other Mac software, and you  
didn't have to know anything about its inner machinery (a compiler  
turning .tex code into .dvi code, a viewer displaying the .dvi code  
on screen, a printer sending the .dvi code to a printing device, and  
a separate text editor). Similarly, a Textures file appeared just  
like any other Mac file (from MacWrite or MacDraw, say), and not as a  
collection of files with naming restrictions (no space, no non-ASCII  
characters) and different extensions.

Sure, over the years Textures started to lag behind, and TeXShop/ 
gwTeX in particular provides now a similar experience with more  
functionality (than Classic Textures) and at no cost. But, because  
Textures invented it all, I will always feel loyal to it and support  
it for as long at it will last (and for how long I'll be able to pay  
for it). That's similar to Mac OS: Mac OS brought the GUI to the  
personal computer, but over the years it started to lag behind and  
became less powerful and reliable than Windows; but, because Mac OS  
invented it all and Windows started as a blatant rip-off of its  
interface, I've always stayed loyal to it, even in the dark days of  
OS 8 and 9 when Mac OS had become slow, messy and unreliable. OS X  
changed all this, of course.


Cons
====

- We're living in a Unicode world; HTML, MathML, XML and all their  
derivatives rule the world. I'm not sure Textures will be able to  
catch up with this evolution. XeTeX, on the other hand, allows this.  
It allows presently to use Unicode fonts as text fonts, and in its  
forthcoming version 0.995 it will allow the use of Unicode fonts (in  
particular the forthcoming STIX fonts) as math fonts. Thanks to  
ConTeXt (and possibly LaTeX 3), it might -- I'm not a specialist --  
get bridges for dealing with XML input and produce XML output.

I'm indeed an avid XeTeX user and beta tester. But sometimes I miss  
the simplicity and minimalism of Textures. In a sense, when one has  
just to write a math paper, say, this simplicity and minimalism allow  
to concentrate on the basics, and be more efficient in this way.  
There's a place for both approaches, I guess.

There are, of course, areas in which XeTeX is unsurpassed. I'm only  
guessing here, but from all the posts to the XeTeX mailing list, it  
seems XeTeX has already become an indispensable tool for many  
linguists, for producing critical editions, for writing non- 
alphabetic scripts, for combining horizontal, vertical, left-to-right  
and right-to-left writing, for unleashing the power of OpenType and  
AAT fonts, etc.

- We're living in a multi-platform world. Back in 1994, when editing  
proceedings, having only Textures to process TeX input sent by  
authors from all over the world turned rapidly into a nightmare. For  
example, having to typeset a communication including figures prepared  
in GNUPlot which added PS instructions to be processed in a dvips  
stream. That's when I started to experiment with OzTeX and interact  
with Andrew Trevorrow (a very nice person).

The gwTeX redistribution of TeX brings this to the next level. With  
it, no need, when receiving TeX input from somebody else, to wonder  
first whether it uses some package or file not present in the  
Textures distro, and whether some adaptation of these  packages is  
required first or is even possible. Instead, you know in advance that  
everything is there and will work. O the added comfort!

Alas, with teTeX now dead and with Gerben's request for help at  
creating from TeXLive a teTeX-like tree, things might change in this  
respect.


There are other things I might say, but I think I've been long enough.

Let me just add that I, too, agree with most of the previous post by  
AES in this thread. And, in particular, that I thank Gerben Wierda,  
Dick Koch, Jonathan Kew and Adam Maxwell for their hard work.

Bye for now,

Bruno Voisin------------------------- Info --------------------------
Mac-TeX Website: http://www.esm.psu.edu/mac-tex/
          & FAQ: http://latex.yauh.de/faq/
TeX FAQ: http://www.tex.ac.uk/faq
List Archive: http://tug.org/pipermail/macostex-archives/