September 2010 Archives

Counterpoint -- modeling isn't an art

| 1 Comment | No TrackBacks
Greetings, readers.  This is a response to Fred's post, "The art of modeling," in which he likens modeling to art.

Actually, I don't want to debate the answer to "what is art".  If a band-aid taped to a wall is considered art, then I'm in over my head anyway.

Ecological modeling, however, is not in the band-aid-on-wall category.  Not yet.  As Fred pointed out, there is growing notion in the literature of a connection between ecological modeling and art.  He cites "subjectivity, instinct, and serendipitous success," all of which certainly come into play, as they do in most scientific endeavors.  Indeed, I've noted this connection myself, often comparing our lab's brainstorming sessions to jazz improvisation (maybe I'm giving us too much credit).

While there may be an "art" to the brainstorming process, the modeling process itself should be rigorous and reproducible.  Non-modelers often perceive models as black boxes, and this perception builds a certain mysticism.  The consequence is a vein of criticism that casts models as "seductive simulations" with appealing visuals, but little connection to the real world.

The response to such criticisms should be in how we conduct our modeling practice.  For example, each assumption, simplification, or approximation should be accompanied with a treatment of how it bears on the final outcome.  A casual observer might view the model output as a work of art, but a critical observer would have all the information necessary to pick the model apart and see how it ticks.

And in response to his raccoon picture, here is a resplendent quetzal.

Nick Record, signing off.


About this Archive

This page is an archive of entries from September 2010 listed from newest to oldest.

August 2010 is the previous archive.

October 2010 is the next archive.

Find recent content on the main index or look in the archives to find all content.