The 1998 Ling-TeX Archives

19th February 1998: WORKSHOP RLA2C - Speaker Recognition

From: RLA2C 
Subject: WORKSHOP RLA2C - Speaker Recognition

******************************************************
*   RLA2C - RLA2C - RLA2C - RLA2C  - RLA2C - RLA2C   *  
******************************************************      
       ---------------      --------------
      la Reconnaissance      Speaker 
              du Locuteur      Recognition
                     et ses      and its 
                 Applications      Commercial
                   Commerciales      and Forensic
              et Criminalistiques      Applications
            
                     -----------------                               
               AVIGNON  20-23 avril/april 1998

                  Soutenu / Sponsored by
                  GFCP - SFA - ESCA - IEEE
              
_______________________________________________________________
Avant Programme/Preliminary Program :
http://lia.univ-avignon.fr/RLA2C/Programme-provisoireWEB.html

Renseignements/Information :
http://lia.univ-avignon.fr/RLA2C

Inscription/Registration :
http://lia.univ-avignon.fr/RLA2C/Inscription.html   
        
Hotels :
http://lia.univ-avignon.fr/RLA2C/Booking.html
_______________________________________________________________  

                       Objectifs / Goal
                       ----------------

En    1994,   un   seminaire     In   1994,   a   specialised
specialise      sur       la     workshop     on    Automatic
reconnaisance automatique du     Speaker Recognition was held
locuteur   s'est   tenu    a     in  Martigny  (Switzerland).
Martigny  (Suisse).   Quatre     Four  years later, the  need
ans plus tard, le besoin est     is  felt  for a new overview
ressenti  de faire un  bilan     of the progress in the field
des progres accomplis et des     and  for discussions on  the
potentialites  des  methodes     practical  impact   of   the
actuelles d'un point de  vue     current    state-of-the-art,
pratique.                        from the commercial point of
Un  deuxieme objectif de  ce     view.
colloque  consiste a  mettre     A   second  goal   of   this
en evidence les specificites     workshop is to emphasize the
des             applications     specificities   of   speaker
criminalistiques    de    la     recognition in the  forensic
reconnaissance  du  locuteur     field and to investigate the
et      l'adequation     des     usefulness  of  the  various
differentes methodes pour ce     techniques for this type  of
type d'applications.             application.
                               
                   
                                                  
          Comite Scientifique / Scientific Committee 
          ------------------------------------------

R. Andre-Obrecht        M. Blomberg             F. Bimbot
L.J. Boe                J.F. Bonastre           L. Boves
G. Chollet              D. Duez                 G. Doddington
S. Furui                J.P. Haton              H. Hollien
J. Mason                H. Meloni               H. Ney
D. Reynolds             M. Rossi                A.E. Rosenberg
J. Schoentgen           F. Soong                I. Trancoso
J. Vaissiere
                               

        Comite d'Organisation / Organisation Committee
        ----------------------------------------------

Ce colloque est organise par     This  workshop is  organised
le  Groupe de Travail  " la     by  the Working Group  (GT1)
Caracterisation du  Locuteur     on   "Speaker  and  Language
et  de  la  Langue "  (GT-1)     Characterisation"       from
du       GDR/PRC-CHM (CNRS)      the      GDR/PRC-CHM (CNRS)

                represente par / represented by

                 Jean-Francois Bonastre, LIA     
             Frederic Bimbot, CNRS/ENST & IRISA
                  Regine Andre-Obrecht, IRIT
                     Bernard Teston, LPL


          Lieu du colloque / Location of the workshop
          -------------------------------------------

Ce  seminaire se tiendra  au     The workshop will take place
Centre d'Enseignement et  de     at the Centre d'Enseignement
Recherche en Informatique de     et     de    Recherche    en
l'Universite  d'Avignon   et     Informatique de l'Universite
des    Pays    de   Vaucluse     d'Avignon  et  des  Pays  de
(LIA/CERI).                      Vaucluse (LIA/CERI)
  
                            
       Droits d'inscription (*) / Registration fees (*)
       ------------------------------------------------
 
Universitaire       Academic        2300 F
  Magistrat        Magistrate

Universitaire       Academic        2000 F
   Membre        Member of one
    d'une           of the
   societe        organising
organisatrice    institution

 Etudiant           Student         1500 F

 Etudiant           Student         1200 F
  Membre         Member of one
   d'une             of the
  societe          organising
organisatrice     institution
   
 Industriel          Private        3000 F
                     company
 
 Sponsor (**)      Sponsor (**)     5000 F
 
Stand pour les       Stand for      Gratuit /Free(***)         
demonstrations      exhibitions
  
  
(*)    Incluant le diner conference/Including the conference-
                                                      dinner
(**)   Incluant un logo en couverture/Including a logo on the 
                                                   cover page
(***)  Dans la limite des places disponibles/Limited to the
                                              space available
                                               
                         
                                                               
                  Secretariat / Secretariat        
                  -------------------------        
 
                       Colloque RLA2C                      
              Laboratoire Informatique d'Avignon   
                 Universite d'Avignon (UAPV)
                      CERI - AGROPARC                     
                         BP 1228                       
                  84911 AVIGNON Cedex 9           
                          FRANCE    
                         
                Tel : (+33/0) 4 90 84 35 09 (ou/or 14)       
                Fax : (+33/0) 4 90 84 35 01     
                
               E-mail : RLA2C@univ-avignon.fr                      
              http://lia.univ-avignon.fr/RLA2C                 
                                   

_____________________________________________________________
Secretariat
Colloque RLA2C / Workshop RLA2C 
Laboratoire Informatique d'Avignon (LIA)                     
Universite d'Avignon et des Pays de Vaucluse                  
Centre d'Enseignement et de Recherche en Informatique         
339 ch. des Meinajaries               
BP 1228 
84911 AVIGNON CEDEX 9         
tel : (+33/0) 4 90 84 35 09/14  
fax : (+33/0) 4 90 84 35 01  
E-M : RLA2C@univ-avignon.fr
www : http://lia.univ-avignon.fr/RLA2C

24th February 1998: To stir things up...

From: Brian Burtt 
Subject: To stir things up...

I subscribed to this list about two weeks ago, and since I haven't seen much
traffic, I thought I'd try to get some discussion going.

If all goes well, I will be pursuing graduate study in linguistics in the fall.
I am interested in studying, documenting, and helping to preserve Native
American languages.  Also (to the extent that time permits) I would like to
continue work with Slavic languages.

So...convince me that TeX would be good for these projects.  I'm trying to
compare this to a standard Windows word processor such as Word Perfect.  It
seems like the former excels in handling long documents quickly and cleanly,
while the latter has the advantage of fonts implemented for more languages and
scripts.

If nothing else, I'd like to hear about what kinds of linguistic publications
you are creating using TeX, and why that is your chosen tool.

Thanks,
--Brian Burtt
burttbri@pilot.msu.edu

24th February 1998: Re: To stir things up...

From: Anshuman Pandey 
Subject: Re: To stir things up...

On Tue, 24 Feb 1998, Brian Burtt wrote:

> So...convince me that TeX would be good for these projects.  I'm trying to
> compare this to a standard Windows word processor such as Word Perfect.  It
> seems like the former excels in handling long documents quickly and cleanly,
> while the latter has the advantage of fonts implemented for more languages and
> scripts.

I use TeX because I was disenchanted by the "obesity" and "automated
features" of MS Word. In fact, I still prepare my TeX documents in WP 5.1
for DOS! It is true that there are more fonts available for use with the
Windows "word-processors", however, if you find Postscript versions of any
font, you can use those with TeX. Also, I believe PCTeX supports the use
of TrueType fonts.

> If nothing else, I'd like to hear about what kinds of linguistic publications
> you are creating using TeX, and why that is your chosen tool.

I use TeX because it is the most efficient and practical method of
typesetting documents which require non-Roman scripts, esp. Indic scripts.
Try implementing 5 or 6 different non-Roman scripts with varied 
directionality (left-right, top-bottom) and different encodings in any 
Windows program and you're likely to give up your graduate work. Also, try
typesetting characters with two degrees of top-level accents in addition
to subscript accents in Windows programs.

Also, there exist style packages for formatting your TeX documents
according to your school's disseration style.

There are more; I'm sure others have opinions too...

Regards,
Anshuman Pandey

24th February 1998: Re: To stir things up...

From: "Michael A. Covington, AI Center, U of Georgia" 
Subject: Re: To stir things up...

The way I see it, MS Word is an automated *typewriter* and LaTeX is an
automated *typist*.

That is, MS Word gives you a simulated sheet of paper with the ability to
write on it any way you please.  LaTeX knows how to type a scholarly
document in a conventional format, and it does that for you.

LaTeX is much better at maintaining a consistent format in a large document
or series of documents.  Also, LaTeX deliberately makes it hard to be
erratic in your layout.  You can't just hit Return three times to position
something; you have to decide what kind of thing it is (a quotation? a
program listing? a linguistic example?) and let LaTeX position it the same
way it positions other things of the same type.

So in LaTeX,  you end up thinking about the elements of your paper
(chapters, sections, numbered examples, formulas) while in Word, you think
about how many lines to skip here or there!

24th February 1998: Re: To stir things up...

From: Christina Thiele 
Subject: Re: To stir things up...

Brian Burtt writes:
> 
> I subscribed to this list about two weeks ago, and since I haven't seen much
> traffic, I thought I'd try to get some discussion going.

Yeah ... well ... we kind of go into seclusion so this is a nice
change -- thanks for the nudge ;-)


> If all goes well, I will be pursuing graduate study in linguistics in the fall.
> I am interested in studying, documenting, and helping to preserve Native
> American languages.  Also (to the extent that time permits) I would like to
> continue work with Slavic languages.
> 
> So...convince me that TeX would be good for these projects.  I'm trying to
> compare this to a standard Windows word processor such as Word Perfect.  It
> seems like the former excels in handling long documents quickly and cleanly,
> while the latter has the advantage of fonts implemented for more languages and
> scripts.

There are two things happening when you do these things: the text
editor, and its flexibility (or not), to allow you to quickly move
through your document, fixing things, moving things, copying things
... all the infrastructure that most software gives you. 

The second component is the program that lets you format what you
want and need, with the fonts you want and need. 

So the first decision is: can I get both of those activities in one
program ... or is it worth using one program for each activity. 

OK. So you're going to follow my bias and opt for one program for each
;-) Emacs (of course ;-)) ) for text editing. You can buzz through a
file or files, or bring in files from someone's else's keyboarding
and/or program ... not that you can't do that with WP or MSWord, of
course. 

Now for the really good reasons to use TeX. WP or MSWord have to have
so many additional little font sets added (since it's unlikely you'll
find everything necessary for NAm Indian langs in just one font) that
the coding for them is pure hell. And if more than one computer's in
use, then you have to make damn'd sure that everyone's got the same
fonts, and is using the same encodings. And then around p.123, you'll
find that you really really need a special diacritic -- not a
preassembled jobbie, but a floating diacritic that you can place just
so ... just exactly where it has to go ... 

Enough of the theory. 

Reality. I'm in the process of picking up the pieces on just such a
WP-just-can't-do-this job ... took the authors almost 2 years of going
through office staff and their equipment and fonts from all over
(including SIL's, I think). Not that SIL's fonts weren't good (they
were) -- the keyboarding, though, had been farmed out over at least
two keyboarders, and then the authors were in there (two more
different machines). 

What I finally got was a hardcopy with blanks for unprintables or
incorrect or just plain ugly IPA chars. I got computer files
(non-matching, since the files were passed around more often than new
hardcopies ... which also makes for `interesting' work!) with codings
all over the map, even some that ended up being the same input for
about three (well, ... I think in one case it was at least six)
different chars (importing into ASCII on a UNIXbox revealed just how
many different encodings overlapped in this unfortunate way). Massive
mess. The problem was too many fonts (and stil there were missing that
would have required hand-rendering), too many encoding options inside
WP, and no consistency or care taken. And the worst of it was that any
screen display was incomplete/defective ... and using F3 for Reveal
Codes only show codes, no characters. So if you've got umpteen special
chars., how you gonna remember them all?! No ... it was not a pretty
demonstration of WP's skills in complete phonetics material. But I'm
pretty much clear of that junk now ...

The book? _Sourcebook for Linguistics_, 2nd or 3rd ed. 465 exercises
with phonetics coming out of your ears. A true test of fonts ... not
to mention accuracy ... 


> If nothing else, I'd like to hear about what kinds of linguistic publications
> you are creating using TeX, and why that is your chosen tool.

Other publications? Papers of the Algonquian Conferences, from the
14th (1983) through to the 25th (1994). Canadian Journal of
Linguistics (ongoing since 1984). Linguistics books for OUP (USA) --
tree diagrams, glossed examples, masses of auto-numbering (great for
catching numbering errors in headings, examples, tables ... ).  Lots
of other stuff not in linguistics ...

I went from plain (with in-house macros) to LaTeX (probably around
1990--91) -- and that change alone made a huge difference. If I
wouldn't consider switching from plain before, there's no way I'd ever
leave LaTeX. Much of what I can now do is because of style files and
such that members of this list have devised:

   Chris Manning's cm-lingmacros.sty
   Emma Pease'     tree-dvips.sty
   ipamac.tex and the WSUIPA fonts (I'll upgrade to TIPA someday, Fukui!)

And from beyond, John Lavagnino's endnotes.sty, Timothy van Zandt's
PSTricks; then xdvi, dvips ... ;-) The support base is incredible. 

And I know that no matter what job I'm handed, I don't have to worry
about some new formatting requirement or some new character (esp.
diacritics) -- if I can't find it already available either via this
list and its members, or from CTAN, then someone can write a macro, or
I can do contortions with \llap's and \raise\hbox's and \h- and
\vspace's that'll put stuff exactly where it has to go. ... and then
wrap it all up inside a macro to make keyboarding a snap.

Nope. No comparison at all, I'm afraid  ... ;-))

Length of the document is not at all the issue. Open-ended flexibility
to produce linguistics as we want it, as we need it -- that's why TeX
is the best.

> Thanks,
> --Brian Burtt
> burttbri@pilot.msu.edu
> 

Wow ... talk about gushing!!! ... It's almost embarrasing, if it
weren't all true ;-)))

Christina 

24th February 1998: Re: To stir things up...

From: "Michael A. Covington, AI Center, U of Georgia" 
Subject: Re: To stir things up...

Christina makes some good points...

Here's an additional concept that's very important.

MS Word and Word Perfect originated as software for typing business letters
on a fixed-pitch printer.  Features were gradually added until these are now
fairly full-featured typesetting programs.

TeX-LaTeX originated in Knuth's research project to implement _perfect_
computer typesetting, equal to the best conventional typography.  LaTeX was
then a macro package added to make it easy to use.

So with Word, you approach good typography from below... with TeX you
approach it from above.

Remember, the history of typography is roughly this:
  (1) Fine book printing was invented, and perfected, by the mid-1800s.
  (2) The typewriter came along and _threw away_ many of the useful features
of standard typography, such as italics and justification.
  (3) Word, Word Perfect, etc., have been gradually adding these features
back in, but usually without sufficient real knowledge of how typesetting
was done prior to the typewriter.

24th February 1998: Re: To stir things up...

From: "Daniel M. Albro" 
Subject: Re: To stir things up...

        One thing I've been able to do in LaTeX is automate example
output.  I've been studying a Losso language called Nauwdm, spoken in
Togo, and I've been able to keep the data in a nice flat file
database, then whenever I want I can extract subsets of the data and with
scripts I can automatically typeset them as nicely glossed example
sentences in a LaTeX document, using gb4e.sty and the TIPA fonts.  I
don't have to edit anything at all.  For all I know something like
this is possible with Word, but I doubt it.

                                              - Dan Albro
                                                Graduate Student, UCLA Ling.

24th February 1998: Re: To stir things up...

From: "Harold F. Schiffman" 
Subject: Re: To stir things up...

We are using TeX and LaTeX to typeset an English-Tamil dictionary; as far
as I know, there's no Tamil font (except a true-type font) for Word or if
there is, I can't imagine how it could work to do something as complex as
format a dictionary. 

H. Schiffman

24th February 1998: Conference

From: Jacques Andre 
Subject: Conference

EP98, the seventh International Conference on
Electronic Publishing, Typography, and Document Manipulation
will held at:

          Saint-Malo, France, April 1st-3rd, 1998

in the frame of a Week on Electronic Documents.that includes
four related conferences (EuroTeX, PODDP, RIDT and EP).

Among lectures, some are connected to linguistics:
- Encoding documents and TEI,
- Dictionnary and ISPELL
- Multiple-Alphabet fonts
- Intelligent Paper
- Textual Aesthetics and Information Retrieval
- Automatic Alignment and Multilingual Documents
- Automatic Generation of Hypertexts
- etc.

Futhermore, a set of tutorials are scheduled during all the week, 
including one on TEI (Text encoding Initiative).

More on this week (complete programmes, registrations, hotel booking,
etc.) at:
http://www.irisa.fr/ep98/week.html

-- 
Jacques André
Irisa/Inria-Rennes,   Campus de Beaulieu,  F-35042 Rennes Cedex,  
France
Tél. : +33 2 99 84 73 50,  fax : +33 2 99 84 71 71, email :
jandre@irisa.fr

26th February 1998: How many ways to trim a tree ... ;-)

From: Christina Thiele 
Subject: How many ways to trim a tree ... ;-)

In casting about for an interesting -- yet TeXnically reasonable --
article for TUGboat, the thought has occurred to me to choose some
particular element in linguistics typesetting, and then see how many
different methods and techniques could be used to arrive at the same
end.

Many inputs ... one output, to be catchy ... ;-)

Soooo ... one element, of course, is tree diagrams. Glosses are
another. Or those AVM structure thingies, although I don't recall
every having to do one. Any others? 

Is there anyone interested in talking about this idea at all? We could
do it here on the list, or privately ... 

One approach would be to find an `interesting' tree (or whatever),
print it up, and then we'd take a crack at achieving that output via
our preferred tried-and-true methods?

Since the list has been pretty quiet these past many months, maybe
this is a way to inject some `spring' into it? 

Ch.

P.S. Any bets on how many of these 'methods' (and their style files)
     might conflict with one another if they're all plopped into the
     same file?! 

2nd March 1998: endnotes.sty, how to use ... ?

From: Christina Thiele 
Subject: endnotes.sty, how to use ... ?

I seem to have suddenly gone quite stupid. I can't for the life of me
remember how to run John Lavagnino's endnotes.sty so that the endnotes
all print up together at the end of a book.

I can't even get it to work now on a project that I did two years ago.
And I don't find any notes that said what the trick was. Maybe certain
files have to be run in a certain order? I dunno anymore ... 


In endnotes.sty, there's the following:

     %     ...  and then add something like
     %
     %     \newpage
     %     \begingroup
     %     \parindent 0pt
     %     \parskip 2ex
     %     \def\enotesize{\normalsize}
     %     \theendnotes
     %     \endgroup
     %
     % as the last thing in your document.

     
... which is followed later on by this:


     %  \theendnotes actually prints out the endnotes.

     %  The user may want separate endnotes for each chapter, or a big
     % block of them at the end of the whole document.  As it stands,
     % either will work; you just say \theendnotes wherever you want the
     % endnotes so far to be inserted.  However, you must add
     % \setcounter{endnote}{0} after that if you want subsequent endnotes
     % to start numbering at 1 again.

====

Which seems very straightforward. So I've just devised a little macro
to put all the first bunch of code together (well, all that I wanted):

\newcommand{\NOTES}{\begingroup 
                    \def\enotesize{\small\baselineskip=11pt}
                    \theendnotes
                    \endgroup\par
                    \par}


Now, if I put \NOTES at the end of a chapter with some \endnote's in
it, then the notes will print. [Note: endnotes.sty generates a file
***.ent, which contains the actual text of the endnotes for a given
file.]

But I have \NOTES in a separate file since I want all the e-notes to
print at the end of the book. But done this way, I get nothing after
the title. Here's my notes.tex file:

     \documentstyle[twoside,pageframe]{odowd-l}
     
     \setcounter{page}{1001}
     \pagestyle{myheadings}
     
     \pageframefalse
     \paperheight=9.25in
     
     
     \begin{document}
     
     \OtherTitle{NOTES}
     
     \widelabel=-.5em
     
     \vspace{-.5pc}
     
     \NOTES
     
     \end{document}
     
     %% END OF FILE


========

I can't for the life of me figure out what point I'm missing. I'm
fairly certain it's the size of a barn door, but ... ;-( 

If anyone has an answer -- and wishes to spare me the embarrassment of
not being able to locate said door -- you can e-mail me direct.

Ch.

P.S. pageframe is a nifty little program from Cameron Smith but it's
     not the source of my problems

3rd March 1998: Re: endnotes.sty, how to use ... ?

From: "Harold F. Schiffman" 
Subject: Re: endnotes.sty, how to use ... ?

While we're at it, I'd like to know how to solve another problem
connectged with \endnotes, which is that it when Latex prints them (I
don't try to put them all at the end, so that;s not my problem) it begins
with a \large or maybe even \Large font for Notes, then starts the first
note *immediately* under this, crowded up against the word Notes, with no
line spacing in between.  When I try to edit the *.ent file and introduce
some space in there, it just ignores me.  What can I do to fix this?

Hal Schiffman

3rd March 1998: Re: endnotes.sty, how to use ... ?

From: Christina Thiele 
Subject: Re: endnotes.sty, how to use ... ?

Harold F. Schiffman writes:
> 
> While we're at it, I'd like to know how to solve another problem
> connectged with \endnotes, which is that it when Latex prints them (I
> don't try to put them all at the end, so that;s not my problem) it begins
> with a \large or maybe even \Large font for Notes, then starts the first
> note *immediately* under this, crowded up against the word Notes, with no
> line spacing in between.  When I try to edit the *.ent file and introduce
> some space in there, it just ignores me.  What can I do to fix this?
> 
> Hal Schiffman


Aha! Something I *can* do ;-)) ... except that all of this is working
under 2.09, not 2e yet, so there may be a hitch here and
there. Dunno. But I figured I'd better put a warning on this stuff. 


Here's what I have for one of my journals, which prints endnotes at
the end of each article.

At the end of the article, I input the command \Notes. Here's how
\Notes is def'd:

\newcommand{\Notes}{{\vspace{6truept}
                     \begin{center}
                        \ttr NOTES
                     \end{center}
                     \NOTES                %% see Endnotes section
                     \vspace{6truept}
                     \enotesize}}          %% see Endnotes section


And those little reminder comments lead me to this chunk in my style
file: 

%% b. Endnotes (coding lifted from belmac-l.sty and modified for ESC's
%%    style):

\input endnotes.sty   %% These are John Lavagnino's endnotes.sty
                      %% macros. 

%% IMPORTANT NOTES (based on reviews work for 21,4):
%%   a. \book now resets the endnote counter back to zero, so it's no
%%      longer necessary to give explicit numbers in []. 
%%   b. to put in an endnote with an asterisk, do the following:
%%        \bgroup
%%        \renewcommand{\theendnote}{\fnsymbol{endnote}}
%%        .... use \endnote normally
%%        \egroup
%%

%% Added explicit baselineskip command, which apparently has to have
%%   that blank space after it, in order to be properly applied.
\def\enotesize{\footnotesize\baselineskip=10pt }

%% Modified \enoteformat to suit ESC style:
\def\enoteformat{\rightskip\z@ \leftskip=11.5pt%
                 \noindent%
                 \leavevmode\llap{\nn{\ttr\@theenmark\en}}}

\newcommand{\NOTES}{\begingroup 
                    \theendnotes
                    \endgroup\par
                   \par}

%% Where an endnote has an indented paragraph, use this:
\newcommand{\enoteindent}{\mbox{\hspace{1em}}}

%% An un-numbered endnote for thanks, acknowledgements, etc.:

\def\acknotetext{\@ifnextchar [{\@xendnotenext}%
   {\begingroup\let\protect\noexpand%
      \xdef\@theenmark{\null}\endgroup%
    \@endnotetext}}
 
\def\acknote#1{{\acknotetext{\baselineskip=10pt \leftskip=0pt%
                                 {\footnotesize\noindent #1}%
                              \vskip3truept}%
          }}


=====================

Now, if you want to use the above, you'd have to check that font
calls, indents, vertical and horizontal spacing suits your
requirements, but ... it may be of use to someone. I also make no
claims about any of the above being elegant or proper LaTeX coding. I
just know it works for me under 2.09. 

Ch.

3rd March 1998: Barn door labelled `ENDNOTES' located...

From: Christina Thiele 
Subject: barn door labelled `ENDNOTES' located ... 

[sheepish message follows]

Well ... with the kind assistance of Alexis Dimitriadis, Bob Frank,
and Uli Sauerland, who gently placed the barn door in front of me
... and Alexis who then tried again to get me to see it ...


John Lavagnino's endnotes.sty creates a .ent file to match the file
with the actual \endnote commands in it: chap1.ent, chap2.ent, and so
on.

So when things are set up to print endnotes at the ends of chapters,
there's a matching set of chap1.tex and chap1.ent. Everyone's happy
the file runs, the endnotes print ... but barn door blindness
prevented me from appreciating this relationship.

When endnotes have to appear at the end of a book, then you have to
build a separate run-file (run-endnotes.tex, in this case) so that all
the proper style files and options and other book-level formatting are
used. But the run-file needs to then find the contents of all those
.ent files in a .ent file to match its own name: run-endnotes.ent.

Oh. I'd deleted that one ... when I prepare files for archiving, once
a project's been completed. I delete .aux files, .log files, .ent
files ... ooops! So there was no trace of what I'd managed to do the
last time. Yet I'd clearly managed to do it. No notes around either.

Well, now there are. In my activity logs. In the run-endnotes.tex
file. In my file on tips and tricks ... here on this list ...  ...

%%  3 MAR 98: the following instructions should be retained (by
%%            deleting all *.ent files for archiving purposes, this
%%            crucial, manually created .ent file is also lost).
%%
%%            This run file needs a matching .ent file. Build that
%%            file in the following way:
%%
%%               \input chap1.ent       
%%               \input chap2.ent       
%%               \input chap3.ent       
%%               \input chap6.ent       
%%               \input chap8.ent
%%
%%            Inserting \section or other heading code where
%%            necessary. If style is to have all notes indent like a
%%            regular paragraph, add \indent before \input, since
%%            \section commands always start text with \noindent, and
%%            the first note will start flush left.
%%
%%  


Everything runs perfectly now. I'm happy. My reputation's been a bit
ding'd, but I'll live ... and remember. 

So I'll just cede the floor now to someone else who'd like to talk
about TeX and linguistics and ... hmmmmmmm

Ch.

4th March 1998: Sephardic orthography

From: Javier Bezos 
Subject: Sephardic orthography

Hello,

Is there any Sephardic speaker reading this?

As you probably know, Sephardi (or Ladino) is a Spanish dialect with 
Hebrew
influences but quite readable to Spaniards. Writing is still 
unstandardized:
some people prefer the fonetic one, but other people advocate for the
Spanish standard ortography with the adition of some diacritical marks.

There is a Spanish style for babel and I think it could be a good idea to
provide a Sephardic variant, but I'm not sure what is the situation 
*outside*
Spain. (Here, as you may guess, the main current opinion is in favour of 
the
Castillian ortography.)

Could someone give me his viewpoint on that?

Thanks
Javier

4th March 1998: xdvi.bin for sunox 4.1.3?

From: Christina Thiele 
Subject: xdvi.bin for sunox 4.1.3?

Subject says it all. 

We're looking for this program to complete the installation of the
TeXLive (v.2) CD materials.

If anyone has a lead, please pass it on to me directly.

Thanks very much.


Christina       

cthiele@ccs.carleton.ca

5th March 1998: Re: TeX Live 2 ;xdvi.bin for sunox 4.1.3?

From: Rama Porrat 
Subject: Re: TeX Live 2 ;xdvi.bin for sunox 4.1.3?

> We're looking for this program to complete the installation of the
> TeXLive (v.2) CD materials.
Speaking about TeX Live2, I have problems installing to an AIX 4.2 .
I used aix 4.1.1 and there has been a number of problems. 
Some of the programs won't run. Does someone have experience with
installin tex on aix 4.2?

I have found out that I must have super user rights in order to
install from the cd rom. This is not mentioned in the documentation,
neither is it convenient. The supervisor is reluctant to give me
such rights. Please refer to this issue in the next version of the
documentation.

Thanks -- Rama.

7th March 1998: Alphabets, linear scripts, cuneiform

From: "Bobby D. Bryant" 
Subject: Alphabets, linear scripts, cuneiform.

I am just starting with LaTeX (and LyX), and would like to get my hands
on fonts and LaTeX support for any/all of the following, if available:

1) Classical Greek, including accents, breathing marks, and the
traditional punctuation marks,

2) epichoric and epigraphic variants of same,

3) the Cypriote syllabary,

4) the Mycenaean Linear B script and ideograms,

5) the Linear A script,

6) the (reconstructed) Phoenician system of signs,

7) and cuneiform suitable for Hittite.

8)  For virtually all the above, the ability to place a dot underneath
the sign to indicate uncertain readings.

9) Naturally, I need an IPA font for phonetic/phonological
transcriptions, so any suggestions as to the "best" packages will be
appreciated.

10) If by chance the community has standardized macros for expressing
sound changes and etymologies, I'd like to adopt them rather than using
an ad-hoc scheme.

Though I have tried to impose a logical order on the above, my most
pressing need is actually for #4, along with enough of #1 (with digamma)
for conventionalized and #9 for phonetic/phonological transcriptions.

With respect to #1, a Web search has turned up GreekTex 3.1, but I raise
the question anyway because the search also turned up an alpha-test kit
v. 4.0 dated 1994, with nothing to indicate its current status.  Any
clarification will be appreciated.  CTAN has a number of other things
tagged "Greek", but the labeling doesn't give much clue as to what they
are, so any suggestions on what I should fetch and play with will also
be appreciated.

---
I am still quite amateur at LaTeX and related tools, but if anyone has
experimental versions of the above I will be glad to try them out and
provide some feedback.  I may prove competent for working on
documentation for them as well.  If you support or are thinking about
undertaking implementations of any of these, please contact me about
participating, but understand that I will be of quite limited use over
the short term.  (However, I'm a big fan of learning-by-doing.)

I will probably provide LyX definitions for any of these that I am able
to obtain, so interested parties may want to contact me w.r.t. that as
well.

---
I am not yet familiar with the conventions for this list.  If it seems
preferable to respond to me off list, I will undertake to post a summary
in a week or two when the probability of further responses has
diminished.

Bobby Bryant
Austin, Texas

10th March 1998: Scribe -> LaTex conversion mode?

From: Avery Andrews 
Subject: Scribe -> LaTex conversion mode?

Has anyone ever encountered a scribe->latex conversion tool, such as an
emacs mode?  I have a big job of that nature to do, and something like
that would certainly be helpful, but this is an inconvenient time for
me to learn significant emacs programming ...

  Avery.Andrews@anu.edu.au

10th March 1998: Re: Scribe -> LaTex conversion mode?

From: "Christopher D. Manning" 
Subject: Scribe -> LaTex conversion mode?

On 10 March 1998, Avery Andrews wrote:
 > Has anyone ever encountered a scribe->latex conversion tool

Avery,

There is a scribe to latex converter.  See:

	ftp://ctan.unsw.edu.au/tex-archive/support/s2latex/

(or people overseas might prefer a nearer CTAN site).

I've never tried it so I don't know how good it is.  If that failed, I'd
try perl rather than emacs.

Chris.

27th April 1998: Spam-filter installed in Ling-TeX list

From: Dag Langmyhr 
Subject: Spam-filter installed in Ling-TeX list

In the past month two spams have been sent to this list. Since this is
often a sign that the name of our list is now in some spanning
database, I have installed a filter to try to stop the spamming.

The filter is based on the idea that only list members may send to the
list. If you have problems, please notify me.

This is my first attempt with this kind of filter, so I may have made
some mistakes. If you get this message, but not the one from `bruker'
(my test account) that I will send 5 minutes after this one,
everything should be OK. Again, if there are problems, please tell me.

						Dag Langmyhr
						(list maintainer)

29th May 1998: Looking for an honest reference...

From: Christina Thiele 
Subject: looking for an honest reference ... 

Anybody for Trivial Pursuit?

A propos of nothing at all ... but these two lists have some of the
finest, most erudite collectors of truly tremendous and occasionally
triumphantly trivial knowledge ...  ...


Story goes like this: a general agrees not to demolish a town if
they'll allow him to take one row away from their protective
wall. Agreed. So he takes the bottom row away ...


To me, the story seems vaguely vaguely familiar ... the question is:
where the heck does this story come from? I'm pretty sure it's not a
Klingon legend ;-)

But just what is the reference? I mean -- what old story or book is
this from? Who wrote it? And most important, what's the name of the
devilishly clever general?

Ch.

3rd June 1998: .vf instead of .mf

From: l.lagerwerf@wmw.utwente.nl
Subject: .vf instead of .mf

Dear TeX-users,

I want to use an adobe  psfont called Stone Serif. The LaTeX support 
files are available at ftp.dante.de. 

I have installed the files in a Windows95 implementation of LaTeX, 
and alternatively on the working directory of my UNIX account. In 
both cases, it doesn't work. The problem is the use of .vf files 
instead of .mf files for METAFONT.

On the PC I forced METAFONT to use .vf files, which resulted in a 
continuous error of illegal characters.

On the UNIX account, I could not force METAFONT into using the .vf 
files. (giving dvips the option -Ppst should do the trick, but it 
doesn't).

Does anyone know a solution to one of these problems?

Kind regards,

Luuk Lagerwerf
Toegepaste Taalkunde Universiteit Twente
053-4893567

3rd June 1998: Re: .vf instead of .mf

From: Barbara Beeton 
Subject: Re: .vf instead of .mf

dear luuk,
stone serif is a commercial font in (postscript) type 1 form; the
shapes are available only by license from adobe.  there is no
metafont source for this font.

a .vf file is a tex convention to make available a non-metafont
font for use with tex.  tex itself uses only the metrics for a
font; it is the output device driver that uses the shapes.

the files you found on ctan (at ftp.dante.de) are the metrics and
the other commands (probably in an .fd file) to make them accessible
to latex.  they won't include the fonts/shapes themselves, because
of the proprietary nature of the font.

so the solution to your problem is to purchase a license from
adobe to use the font.
						-- barbara beeton

7th July 1998: Optimality theory

From: Johannes Reese 
Subject: Optimality Theory

Hi,
I want to produce a Optimality Theory tabular with LaTeX; the problem is
about the gray fields for the alternatives that are not chosen. I tried
the psboxit-package, but first it only coloured a quarter of the box,
second it didn't work inside a tabular-environment.
Does anyone know the solution?
Johannes

7th July 1998: Re: Optimality theory

From: Christina Thiele 
Subject: Re: Optimality Theory

Johannes Reese writes:

> Hi,
> I want to produce a Optimality Theory tabular with LaTeX; the problem is
> about the gray fields for the alternatives that are not chosen. I tried
> the psboxit-package, but first it only coloured a quarter of the box,
> second it didn't work inside a tabular-environment.
> Does anyone know the solution?
> Johannes


Can I ask if this is what you mean?

You have a table, with rows and columns, with vertical and horizontal
lines separating each `cell'?

If you happen to have a copy of issue 41,4 of the Canadian Journal of
Linguistics, would it be like the grey cells in Doug Pulleyblank's
article, ``Neutral Vowels in Optimality Theory ... ''' euh ... hmmm
... looks like it ;-)

OK. That answers that ... here's how I did the tables.

1. get both pstricks *and* colortab from CTAN -- unless you've already
   got them to hand -- and load them as options to the documentstyle
   ... yes ... I know, I'm still on 2.09, so it's \documentstyle. But
   I'll bet it all works in 2e with \documentclass ... 

2. I reset rule widths so that everything was the same (since I also
   use Emma Pease' tree-dvips and the tables had mini-trees in 'em):

   \treelinewidth=.4pt   %% default = .3pt
   \arrayrulewidth=.4pt  %% rules inside shaded boxes same as other rules

3. I chose the following level of grey:

   %% PSTricks parameter:
   \newgray{lightergray}{.90}  %% ligher than default \lightgray value

4. Here's the coding for ex.11 in CJL 41,1 (p.301):

%% Ex 11:
\Example{{\it Default advancement} 

\font\ipa=wsuipa8
\renewcommand{\arraystretch}{.8}
\begin{tabular}{|lcc|c|c|}
\mc{1}{l}{\empty} & \\ [-12pt]
\cline{4-5}
\mc{1}{l}{\empty} 
    & \mc{1}{l}{/EgE/} & \mc{1}{c|}{`dirge'} & \sc DepRtr & \sc DepAtr\T\B \\
\hline
\LCC
   &&  & & \lightergray \\
a. && \fns \tabcolsep=1pt

      \begin{tabular}{ccc}
          \mc{3}{c}{\node{a}{RTR}}\T \\ [1ex]
          \node{b}{\niepsilon} & g & \node{c}{\niepsilon} \\
      \end{tabular} & *! & \\
\hline 
b. & \hand
    & \fns \tabcolsep=1pt
      \begin{tabular}{ccc}
          \mc{3}{c}{\node{d}{ATR}}\T  \\ [1ex]
          \node{e}{e} & g & \node{f}{e} \\
      \end{tabular} & & * \\
\hline 
\ECC
\end{tabular}
%  \node{}{}
\nodeconnect{a}{b}  \nodeconnect{a}{c}  \nodebox{a}  
\nodeconnect{d}{e}  \nodeconnect{d}{f}  \nodebox{d}
}


Notes to all that code ;-))

1. I use a variant of enumsentence from Chris Mannings cm-lingmacros,
   and call it \Example

2. I reset the IPA to be 9pt inside the example.

3. The \arraystretch is reset to make things look less gappy.

4. \mc = \multicolumn ... to save space.

5. \LCC starts a grey row, \ECC turns it off. Notice that essentially
   you print a grey row, then print a text row, which sits ?on top/
   ?below the grey field. 

===

I have the PStricks manual printed up and handy in a binder. But I
don't see now where I found the reference to colortab but I believe I
picked it up from CTAN along with the documentation, also printed up. 
Keep 'em both handy -- this is great stuff for all kinds of nifty
one-time things. 

Good luck!

Ch.

7th July 1998: Re: Optimality theory

From: Christina Thiele 
Subject: Re: Optimality Theory

Forgot to include these:

1. to reduce font size inside the tabular which is inside the larger
   tabular: 

   \let\fns=\footnotesize

2. in Doug's article, I had to find a `pointing finger', which I got
   from the Zapf Dingbat font:

   \font\speciala=pzdr at 12pt
   \newcommand{\hand}{\rlap{\speciala\char'053}}

3. also, I originally tried LaTeX's framebox, I think it was, to 
   put a box around elements inside the `tree' things inside the
   tables, but it was too awkward ... then I noticed \nodebox in
   Emma's documentation for tree-dvips and that made things a lot
   simpler

Ch.

20th July 1998: Web site

From: "Geert-Jan M. Kruijff" 
Subject: Website 

FYI,

We are maintaining a repository of (links to) "LaTeX packages relevant for
linguists" here at the Institute  of Formal and Applied Linguistics. The
repository is reachable via the institute's main page,

		http://ufal.mff.cuni.cz

Packages are mainly intended for use with LaTeX2e. Most of the time,
documentation on how to use a specific package is available.

Cordially, Geert-Jan Kruijff

-----------------------------------------------------------------------
Geert-Jan M. Kruijff
Institute of Formal & Applied Linguistics/Linguistic Data Laboratory
Faculty of Mathematics and Physics, Charles University

Malostranske nam. 25, CZ-118 00 Prague 1, Czech Republic
Phone: ++420-2-2191-4255 Fax: ++420-2-2191-4309
Email: gj@ufal.mff.cuni.cz, gj@acm.org
WWW:   http://kwetal.ms.mff.cuni.cz/~gj/

8th September 1998: Re: Linguistic structures side by side

From: away@COMPAPP.DCU.IE (Andy Way CA)
Subject: linguistic structures side by side

I need to draw a PS-tree next to an A-V matrix,
and draw lines between various nodes in the tree
and their corresponding nodes in the array, that
is, relate explicitly c-structure nodes with their
f-structure equivalents in LFG.

I know how to do all these things in isolation
using various macros (Emma Pease's are what I use),
but not side by side in the same diagram. Here's
a (crappy) visualisation of what I need:


	S			 _		__
       /  \			|SUBJ	'Kim'	  |
       NP  VP			|TENSE   PRES     |
	|   |		        |PRED eats<^SUBJ> |
       Kim  eats		 -		 -

with lines (arrows) drawn betw:
	- the {NP,Kim} nodes and the 'Kim' part of the f-struc.
	- the {S,VP,eats} nodes and the outer f-struc.

Any clever people out there who can do this? Otherwise I'll
have to use xfig ....

Cheers,
Andy.

8th September 1998: Linguistic structures side by side

From: Emma Pease 
Subject: Re: linguistic structures side by side 

> I need to draw a PS-tree next to an A-V matrix,
> and draw lines between various nodes in the tree
> and their corresponding nodes in the array, that
> is, relate explicitly c-structure nodes with their
> f-structure equivalents in LFG.

Assuming you have the right packages, you should just be able to
combine them.

For instance


\usepackage{lingmacros} % csli lingmacros macros
\usepackage{avm} % Chris Manning's avm macros
\usepackage{tree-dvips} % Emma's node drawing macros

...

\modsmalltree{3}{&\node{a}{S}\\
\node{b}{NP} && \node{c}{VP}\\
\node{d}{Kim} && \node{e}{eats}}
%
\hspace{1in}
%
\node{f}{
\begin{avm}
\[ SUBJ & \node{b1}{'Kim'}\\
   TENSE & PRES\\
   PRED & eats[SUBJ] \]
\end{avm}}

\nodeconnect{a}{b}
\nodeconnect{a}{c}
\nodeconnect{b}{d}
\nodeconnect{c}{e}

\nodecurve[r]{b}{b1}{.4in}
\nodecurve[r]{d}{b1}{.4in}
\nodecurve[r]{a}[l]{f}{.4in}
\nodecurve[r]{c}[l]{f}{.4in}
\nodecurve[r]{e}[l]{f}{.4in}

****

Emma

9th September 1998: Linguistic structures side by side

From: Dirk Kussin 
Subject: Re: linguistic structures side by side

I think everything is possible with the xypic package. I attach
information about it at the end of this mail. Good luck.

Dirk

   I need to draw a PS-tree next to an A-V matrix,
   and draw lines between various nodes in the tree
   and their corresponding nodes in the array, that
   is, relate explicitly c-structure nodes with their
   f-structure equivalents in LFG.

   I know how to do all these things in isolation
   using various macros (Emma Pease's are what I use),
   but not side by side in the same diagram. Here's
   a (crappy) visualisation of what I need:


	   S			 _		__
	  /  \			|SUBJ	'Kim'	  |
	  NP  VP			|TENSE   PRES     |
	   |   |		        |PRED eats<^SUBJ> |
	  Kim  eats		 -		 -

   with lines (arrows) drawn betw:
	   - the {NP,Kim} nodes and the 'Kim' part of the f-struc.
	   - the {S,VP,eats} nodes and the outer f-struc.

   Any clever people out there who can do this? Otherwise I'll
   have to use xfig ....

   Cheers,
   Andy.

-- 
Dirk Kussin                 dirk@uni-paderborn.de     
Fachbereich 17 Mathematik   Raum D2.323
Universität-GH Paderborn    Tel. (+49) (5251) 60-2636
D-33095 Paderborn --------- http://www-math.uni-paderborn.de/~dirk/


--------------------------------------------------------------------
>From krisrose@sauternes.ens-lyon.fr Fri Mar  6 05:59 MET 1998
Date: Fri, 6 Mar 1998 06:05:22 +0100 (CET)
From: krisrose@sauternes.ens-lyon.fr (Kristoffer Rose)
To: dirk@uni-paderborn.de
Subject: ANNOUNCING: Xy-pic version 3.6 released!
Content-Type: text
Content-Length: 4696

Dear Dirk Kussin,

Please find enclosed a copy of the TRAILER for a new version of Xy-pic!

Sincerely,
           Kristoffer H. Rose

=======================================================================
    ANNOUNCING the Xy-pic version 3.6 DIAGRAM TYPESETTING PACKAGE
=======================================================================

This is to announce a release of my diagram typesetting package Xy-pic.

Version 3.6 contains new and improved PostScript fonts donated by Y&Y
Inc. (as well as several minor bug fixes).

-----------------------------------------------------------------------
				GENERAL
-----------------------------------------------------------------------

Xy-pic is a package for typesetting a variety of graphs and diagrams
with TeX.  Xy-pic works with most formats (including LaTeX, AMS-LaTeX,
AMS-TeX, and plain TeX), in particular Xy-pic is provided as a LaTeX2e
`supported package' (following the `CTAN LaTeX2e bundle' standard).

Further specifics of the package are in the distribution README file.

-----------------------------------------------------------------------
				 NEWS
-----------------------------------------------------------------------

The primary purpose of release 3.6 is the inclusion of much improved
PostSript fonts courtesy of Y&Y TeX and several Internetwide
volunteers: these make it possible to use Xy-pic output with Windows
(in addition to Macintosh and Un*x), Adobe's ATM software, and to
produce Adobe PDF ("Acrobat") files.

NOTE: This change means that you must REMOVE SOME OLD FONT FILES (the
details of which and how are in the INSTALL file).

Release 3.6 furthermore fixes a few problems of release 3.5:

* frames more robust against disappearing on bogus PostScript printers
* thickness of curves now adjustable
* TDS compliance improved
* bogus `graph escape redefined' message eliminated

Thanks to David Carlisle, John Landamore, Ralph Loader, Thorsten
Schwander, and Timco Visser, for your bug reports, and to Adam
Worrall, Thierry Bouche, Tanmoy Bhattacharya and especially to
Berthold Horn and Louis Vosloo of Y&Y Inc. for their work on the
PostScript fonts.

Finally profound thanks to BRICS and DAIMI for hosting Xy-pic (and
Kristoffer :) for a most productive and enjoyable year!

-----------------------------------------------------------------------
			      AVAILABILITY
-----------------------------------------------------------------------

Xy-pic can be retrieved through the World Wide Web Xy-pic `home pages':

  URL:			http://www.ens-lyon.fr/~krisrose/Xy-pic.html
  URL:			http://www.mpce.mq.edu.au/~ross/Xy-pic.html

as well as by anonymous ftp from

  CTAN:			macros/generic/diagrams/xypic

and from the private TeX archives of the authors:

  ftp.ens-lyon.fr : 	/pub/users/LIP/krisrose/TeX/
  ftp.mpce.mq.edu.au :	/pub/maths/TeX/

Check the README file in each location for specifics, in particular
check that you have reached a version 3.6 copy (some archives take a
while to mirror the latest files)!

-----------------------------------------------------------------------
			   HISTORY & CREDITS
-----------------------------------------------------------------------

The first public release (version 1.40) of Xy-pic was created by
Kristoffer H. Rose, DIKU, U of Copenhagen (now ENS-Lyon, France) and
distributed via Usenet on December 19, 1991.  This quickly became
version 2 of which version 2.6 was the most stable.

The thorough rewrite that became version 3 is a continued
collaboration with Ross Moore, Macquarie U, Sydney, initiated through
a visit to Macquarie (Jan-May 1994 supported by the Australian
Research Council, Macquarie University, and using donated DEC
equipment).  However, full backwards compatibility is maintained
(except for the unavoidable but fully documented obscure cases).

Xy-pic is Copyright (c) 1991-1998 by Kristoffer H. Rose and 1994-1998
by Ross Moore under GNU COPYLEFT which means that you can use the
package for any purpose but if you provide the macros or any code
derived from them to a third party then you are obliged to include the
entire Xy-pic package (full details in the file COPYING).  The
PostScript fonts are additionally Copyright (c) 1997 Y&Y Inc.

----------------------------------------------------

	 This is the end of the announcement.  Enjoy Xy-pic!

-----------------------------------------------------------------------

Kristoffer Høgsbro Rose, Ph.D., prof.associé  
Laboratoire de l'Informatique du Parallélisme  équipe PLUME, bureau LR5-026
Ecole Normale Supérieure de Lyon; 46, Allée d'Italie; F-69364 Lyon 07 cedex
phone: +33(0)4 7272 8642; fax:...8080    

Last updated 1998/12/07 by Dag Langmyhr.