Force-directed layouts become difficult to interpret if there are more than 
a hundred nodes or so, or if there are a lot of multi-way relationships. A 
large number of generally hierarchical nodes (such as In many directed 
acyclic graphs) is often easier to interpret using a hierarchical layout 
(as described in the OP's linked paper). A graph with numerous multi-way 
relationships can be easier to interpret with a chord diagram or similar.

A force-directed layout that allows you to drag and "pin" nodes could be 
made to work, but then you're just being forced to do manually what the 
computer ought to be able to do for you.

Here's the central portion of a complex DAG I happen to be working with 
right now, rendered as a force-directed graph in D3: 
http://imgur.com/a/lBP6i

The nodes are actually almost strictly hierarchical, but you'd never know 
that looking at this image.


On Saturday, August 5, 2017 at 7:09:34 PM UTC-4, Jakub Hampl wrote:
>
> In principle this sort of thing would be in scope for elm-visualisation to 
> implement. However, as that particular approach entails a lot of work, I 
> would like to hear about the details of the use case and whether the force 
> layout approach could not be adapted to work. 

-- 
You received this message because you are subscribed to the Google Groups "Elm 
Discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to elm-discuss+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to