[TriPython] RTA All Graphs Are Wrong, But Some Are Useful at RENCI April 19
Chris Calloway
cbc at unc.edu
Fri Apr 1 15:04:03 EDT 2016
On 4/1/16 12:14 PM, Francois Dion wrote:
> Clever repurposing of George Box's "all models are wrong [...] how wrong
> do they have to be to not be useful". I seem to remember that John Tukey
> had a similar saying on the display of data.
Yeah, Xan Gregg references John Tukey by name in this talk.
> On the use of "graph" to talk about data-based displays, I don't know that
> it is weird from a stats, operations research or graphical design
> background. It is more of an issue from a discrete math or computing
> science where we tend to think in terms of graph = networks.
All of which are in the realm of "data science." Kind of like how the
word "postmodern" means entirely different things in different branches
of arts and humanities. Between choreography and architecture, almost
opposite meanings (cue Frank Zappa quote).
--
Sincerely,
Chris Calloway, Applications Analyst
UNC Renaissance Computing Institute
100 Europa Drive, Suite 540, Chapel Hill, NC 27517
(919) 599-3530
More information about the TriZPUG
mailing list