In The Armchair

LaTeX Auto-Preview

Posted in Computers by Armchair Guy on September 24, 2009


It never ceases to amaze me that one or two brilliant people working by themselves can sometimes invent something that is so far superior to a professional piece of software produced by a behemoth like Microsoft.  Knuth created TeX, Leslie Lamport evolved it by creating LaTeX, and we keep getting huge improvements (like PGF/TiKZ) from time to time.  I find myself very grateful for some of the other things available for LaTeX, one of the most important being auto-preview.

There are three main auto-preview methods for LaTeX while editing under emacs.  These are collected at

WhizzyTeX is the most evolved.  It continuously updates a separate dvi-preview window as you type in emacs, and also positions the dvi preview at the point where you are editing.  The biggest problem with whizzytex is that it doesn’t understand newer packages like PGF/TiKZ which directly compile to PDF using pdflatex.  Such documents don’t compile to DVI at all, so whizzytex is no good here.

LaTeXMk is essentially a ‘make’ for LaTeX but it can auto-update every time you save.  So it won’t update continuously but will refresh as soon as you save (C-x C-s) under emacs.  This will also work with any other editor; it follows filesystem level changes to the file.  But:

  • It does not position the pdf viewer at the point you edited; you have to keep doing this yourself.
  • When there’s an error in compiling LaTeX code,  LaTeXMk calls LaTeX, which stops on the error.  You have to manually go into the LaTeXMk window and quit LaTeX for it to continue. (Correcting the error within emacs isn’t enough.)

Rubber appears to be similar to LaTeXMk.  I’ll update once I’ve given it a whirl…

Other that these, there are some What You See Is What You Mean programs like Lyx and Scientific Workplace (commercial).  Both are excellent, but sometimes it’s hard for me to get used to the lack of control.

India’s Genetic History

Posted in India by Armchair Guy on September 24, 2009


A Nature article partially tracing the genetic history of Indian populations was published September 24:

Original Nature article (abstract): Reconstructing Indian Population History – Reich et. al. – 2009

An HSPH news release: New Research Reveals the Ancestral Populations of India and Their Relationships to Modern Groups

Sounds like rich fodder for the Aryan Invasion Theory debate.  But the more serious consequence has good news and bad news: quoting from the abstract,

there will be an excess of recessive diseases in India, which should be possible to screen and map genetically

I thought the first line of the abstract is pretty important too:

India has been underrepresented in genome-wide surveys of human variation.

Hopefully said underrepresentation will cease!