I read a couple of inspirational blogs recently:
- Bora Beran’s coxcomb chart – http://boraberan.wordpress.com/2014/03/30/creating-coxcomb-charts-in-tableau/
- RedHeadedStepData’s post on mastering Tableau concepts – http://redheadedstepdata.io/master-tableau-concepts/
These both discussed the use of domain padding and data densification to extrapolate or fill-in on data that is not included in the data source. Bora’s application of this capability to interpolating curves inspired me to revisit a workbook I had made a couple of years ago showing mobile tower transmitters. At the time I simply triple-read the input data using a UNION and I ended up with triangles showing the sectors:
What I had really wanted was curved sector arcs but I couldn’t scale the solution to produce a smooth curve. Applying Bora’s technique I was able to rectify this easily and plot with as many interpolated points as I want using only two passes over the data:
See these workbooks for the before/after approaches.
This then got me thinking about other applications this technique could be used for and the first “curved line” solution that came to mind was plotting great circle routes on maps. I’ve been asked if Tableau could do this in the past and the answer was no, but using this approach I produced the following:
I densified the data to produce points for interpolation and then passed the data into R. There I used the gcIntermediate function to generate the intermediate lat/long points and plotted the resulting line. See this workbook for the solution.
I think this same technique could be used for a number of other interesting applications such as Sankey diagrams. Hopefully you find it useful.