Samples as experiments

Today’s tidbit is about sampling.  We are often told to “sample, sample, sample” but not given a solid reason why or how to sample.  Or rather, we are given only part of the answer: one weaves samples to see what the finished cloth will look like.

I’d like to offer a more expansive definition of sampling:

A sample is an experiment  to answer a question about design or execution of a project.

This is very different from “put an extra half-yard to yard of warp on the loom to experiment with”.  In my world, woven samples are only half the story, and the most expensive half!

My feeling is that an efficient sample is one that answers the design or execution question with minimum effort.  For design questions, an initial sample might be a sketch of the project.  This provides a quick way to look at overall composition before launching into more detailed work.  Once you are satisfied with the quick sketch, more detailed simulations might be effective; the exact nature of the simulation depends on the question that you are trying to answer.

In essence, a sample is an experiment – it answers the question, Does this work?  “This” might be an aspect of design or composition, or it might be a question about construction.  The most important part of sampling, in my opinion, is to know the question you are trying to answer and design the sample to answer that question as efficiently as possible.  Wondering about color in a garment?  Whip out Photoshop, or your colored pencils, and do a mockup.  If you still aren’t sure, dye some cheap fabric in your proposed color combinations and sew it into a muslin.  But consider the cost of your samples when deciding how to sample: the dyed-muslin sample is much more expensive – both in time and in materials – than a ten-minute Photoshop exercise, so if Photoshop will do, skip the muslin.  The idea is to execute your experiments with a minimum of investment, to answer your questions most efficiently.