David Parker wrote:
> First, let me congratulate you all on the beautiful work that you all have 
> done.
> The graphical layout is both pleasing to the eye as well as informative.

Thank you.

> I've have been thinking about how to add more dimensions to a 2S 
> representation

<sigh/>

> 1. Could the color of an individual line change from red to blue through the 
> spectrum according to a preset maximum and minimum range?  This might also 
> apply to its luminosity or transparency.

Hmm, no.

It's entirely possible to modify timeplot to allow the painting function
to be pluggable but right now it's not, it's fixed. I could be convinced
to do that but I would need at least 3 different people asking for it
and with a substantial need for it... and without negatively impacting
the existing user base.

> 2. Could the width or thickness of an individual line also change according 
> to scale that is a relationship with a preset maximum and minimum range?

With the above you could do the line painting in any way you like.

> For example,  Napoleon's Russian Campaign by Count Philippe-Paul de Ségur has 
> the x-coordinate represent both time  and location (Paris to Moscow). The 
> width being the size of the army. But he added an extra plot line for the 
> temperature. This could have been done with a color component.

Uh, aiming pretty high there, aren't you ;-) (that plot is considered by
Tufte the best plot ever).

The problem of plotting a (t,v1,v2) time series is, IMO, radically
different than plotting a (t,v) time series, unless, of course, one
could write the time series as (t,v,f(v)) which Timeplot already supports.

The use of color coding or shape coding for v2 is a good suggestion but
I fear the complexity of going down that path (for example, one could
expect (t,v1,v2) to become (t,v1,f(v1),v2,g(v2)) and so on.

Color coding on such a thin line is probably not the best solution and
shape coding could be perceived either as a visual artifact for small
values or as very distorting for bigger values.

I don't know, I'm not entirely against it but right now I fail to see
real value for it. Can you elaborate more on a few practical scenarios
where you would like to use this?

-- 
Stefano Mazzocchi
Digital Libraries Research Group                 Research Scientist
Massachusetts Institute of Technology
E25-131, 77 Massachusetts Ave               skype: stefanomazzocchi
Cambridge, MA  02139-4307, USA         email: stefanom at mit . edu
-------------------------------------------------------------------

_______________________________________________
General mailing list
[email protected]
http://simile.mit.edu/mailman/listinfo/general

Reply via email to