[pracjourn-forum] A weird error (and broader issues)

[Peter Flom]

Thanks to all who replied.  

Karl wrote, in part

<<<
When I run your document with latex, it works fine for me.
When I run your document with pdflatex, I get the error above.
>>>

This is definitely something to keep in mind, and I haven't seen it
mentioned anywhere.  That is, people should be told 'if pdfLaTeX doesn't
work, try LaTeX".   

A related note:  with proTeXt, there's a button pdfTeXify.   This saves
time by running pdfLaTeX multiple times (to resolve bibliographies and
TOC and such) but bogs down if there are problems with the file.

<<<
I cannot quite imagine any rational way that you could glean
information
from the error message or log file to determine this.  The only viable
answer I can think of is for pstricks to detect that it is running
under
pdflatex (which is doable) and give a meaningful error.  I will write
Herbert Voss about it.
>>>

I don't know, but it might be that this happens with more than just
pstricks.  I know Ive gotten similar errors, but am not sure if they
have been due to pstricks or some other package.



    with undeifined control sequences for things I never typed....

<<<<
Indeed.  The general answer is that the log file lists every file
input
by TeX (unfortunately not in a very convenient form).  If you search
those files for the string in question ("c at lor@to at ps" -- best to leave
off the backslash when searching) usually [though not always] you will
find it, which will at least tell you the package involved.  In this
case, "c at lor" is a clue that color might be involved, although that
doesn't say much.

Unfortunately I don't know of a tool to make this kind of searching
convenient.  I've wanted it myself, more than once.  Surely there's
something out there ...
>>>

I think this actually goes to a larger point, which I made in my recent
article, and plan to stress in my talk.....that debugging is an
essential part of the process, and that teaching people  a) that
debugging is normal and b) how to do it   are essential parts of
teaching people to use LaTeX. 

Coming (as I do) from a background completely removed from programming,
this is not 'natural' to me, and this has both psychological and
practical ramifications in broadening the use of LaTeX


Peter

Peter L. Flom, PhD
Assistant Director, Statistics and Data Analysis Core
Center for Drug Use and HIV Research
National Development and Research Institutes
71 W. 23rd St
www.peterflom.com
New York, NY 10010
(212) 845-4485 (voice)
(917) 438-0894 (fax)