[prev in list] [next in list] [prev in thread] [next in thread] 

List:       graphviz-interest
Subject:    Re: [graphviz-interest] tools to visualize just parts of a graph
From:       "Fabrizio Morbini" <fmorbini () gmail ! com>
Date:       2007-01-22 21:21:23
Message-ID: e0660b1d0701221321k3af27719lbd203fa30aad6149 () mail ! gmail ! com
[Download RAW message or body]

in my case the number of nodes is not overwhelming, but the
connections are kind of intricate and one would like to see just the
connections to and from a particular set of nodes without any other
crossing edge. I saw in action dtistudio from johns hopkins university
and thought that that functionality is what i need. even in the case
of the brain the whole structure is not too big, but the network of
connections is too intricate. This is why, i guess, they thought of
this region-of-interest based display.

thanks for all these suggestions, i'll try dynasty as soon as available.
fabrizio.

On 1/22/07, Gordon Woodhull <gordon@woodhull.com> wrote:
> Dynasty sounds like the tool for you.  We haven't quite released it, but
> I'll try to make that happen soon.  It will be GPL.  It uses Dynagraph
> for layout, which is of course CPL.
>
> I think it's quite usable; it's in its second generation and my
> colleagues at Johns Hopkins are using it for debugging huge NLP proof
> trees.  You can view as many or as few nodes as you want and then
> navigate either using the keyboard or by clicking on new center nodes.
>
> Fabrizio, I'm not sure from your question whether you mean that the
> whole graph should be laid out and then only a small part shown.  This
> is not what Dynasty does; it's intended for cases where the whole graph
> could not be laid out because it's too large.  Therefore it only lays
> out what's on the screen and then when you navigate from there it adds
> nodes and edges, trying to balance stability with readability...
>
> Another minor difference from what you are describing is that Dynasty is
> currently programmed to show a certain radius from a cursor node, rather
> than showing all ancestors or all descendants.  However, I'm sure it
> would be easy to modify its exploration algorithm.
>
> Cheers,
> Gordon
>
> Stephen North wrote:
> > This capability is really what we were aiming at with the
> > Dyngraph system.  Gordon Woodhull and some colleagues
> > from Johns Hopkins had a poster on this style of graph
> > browsing at the last IEEE Infovis conference. I'm ccing
> > him on this message.
> >
> > The code is open source and downloadable
> > but I'm not sure how easy it is to get up and running.
> > Gordon should be able to explain further.
> >
> > Stephen
> >
> >
> Fabrizio Morbini wrote:
> > Hi all, is there some tool that is able to draw just a part of the
> > edges in a big graph based on a region of interested selected by the
> > user?
> > for example, in a tree, the user could select a single node and the
> > tool should draw the edges coming to that node and leaving that node
> > (till leaves or root are reached).
> >
> > thanks, fabrizio.
> > _______________________________________________
> > graphviz-interest@research.att.com
> > https://mailman.research.att.com/mailman/listinfo/graphviz-interest
> >
>
>
_______________________________________________
graphviz-interest@research.att.com
https://mailman.research.att.com/mailman/listinfo/graphviz-interest
[prev in list] [next in list] [prev in thread] [next in thread]