[texhax] Simple graph theory diagrams in a document

Matthew Leingang leingang at math.harvard.edu
Fri Nov 23 17:15:25 CET 2007

On Nov 21, 2007, at 2:30 PM, Tom Backer Johnsen wrote:

> Colleagues:
> Almost exactly a month ago, there were a question about drawing in a
> LatTex document.  I knew I would need a solution to that kind of
> problem, but not just right then.  So, I made a note, and left the
> problem for the time being.  Today, I looked at what I assumed was the
> outcome of the recomendations: pgf/tikz.
> Dear me, a 248-page manual in color.  A very powerful tool.
> Obviously.  And very obviously a blatant overkill in my case.
> I am writing a text within psychology there I employ basic stuff in
> graph theory.  The primary tool is what is called "signed and directed
> graphs", consisting of a set of points ("dots" on the paper) and a set
> of relations between them in the form of arrows (dashed for negative
> and solid for positive, which may be curved if there are arrows in
> both directions between two points).
> So I need to draw something like that in the MS, only consisting of
> very few points, maximum of say five.  Possibly with some kind of
> label (x1, x2 etc.) on the points, and possibly with some value (-2 to
> +2) associated with the relations or arrows.
> In other words, what I need is something that is "good enough", less
> of an overkill, but still producing nice diagrams.
> Any suggestions?
> Tom


You got lots of responses but I want to defend the TikZ option.

There's a very helpful group of users on the mailing list so you're  
not alone with the 248-page manual:


If you're doing a lot of these diagrams, a solution within TeX is  
going to speed up your workflow a lot.  You can copy-and-paste code  
to make new diagrams from old diagrams.  If you decide to change  
styles, you can do it in one fell swoop.  And the language itself is  
very well suited to exactly the kind of diagrams you want to make.   
You can create and name "nodes" and draw arrows between them.


Matthew Leingang
Preceptor in Mathematics
Harvard University


More information about the texhax mailing list