My Photo

« Data Engine Roundup #8 | Main | International Time Series Data Available on d8taplex »

April 10, 2011



Nice to see these type of applied CHI works :)


Without looking at the paper, I can't work out which direction a tapered line is representing. The thinner end could be the direction as it kind of points to something, but the way it fades out confuses me. It looks more like a motion trail to me (picture a commet), where a node is moving away from the other node, which gives it the feeling of having the opposite direction of the first interpretation (in fact, this felt more natural to me).

I'd be interested in hearing other people's thoughts (before you read the paper).


@Alexbowe, agreed. They look like tadpoles. Try drawing fancy tapered lines on a whiteboard when you're trying to explain a directed graph to someone. Arrows may be clunky but at least they're simple and unambiguous. Of course, if an artistic interpretation of the data is the goal then do whatever suits your fancy :)



I'm agreeing with Alex: I'm not sure what part is what. Suppose it's readership, it makes more sense to have the thinner edge on the writer's side—they tend to have more fans then fans have favorite references. If it's payments though, I'm not sure who send to whom.

Robert Renaud

Without reading the paper, I am >90% certain that the line gets fatter in the direction the arrow would point.


Now that I've skimmed the paper, it seems they note the possible ambiguity, but their tests show that people have a faster reaction time.

I guess it might be better if you clearly indicate early on which direction it means.

The comments to this entry are closed.

Twitter Updates

    follow me on Twitter

    March 2016

    Sun Mon Tue Wed Thu Fri Sat
        1 2 3 4 5
    6 7 8 9 10 11 12
    13 14 15 16 17 18 19
    20 21 22 23 24 25 26
    27 28 29 30 31    


    Blog powered by Typepad