Sampling: It’s not what you think

Few topics seem more divisive for weavers than sampling. Some people enjoy sampling and weave a lot of samples, others hate sampling and rarely do it.

I’ve heard many weavers say, “Yes, I really should sample, but I rarely do.” Often, weavers treat sampling as a moral good – that is, a good habit, reflecting discipline, that everyone should do. (And feel guilty about NOT doing.)

That’s not how I view sampling.

What’s a sample?

Samples, as I see it, are simply experiments. You invest a little time and materials to find out whether some aspect of your design will work. Samples don’t have to be woven samples – they can be anything that helps you test your design. A black-and-white photo, a draft simulation, a yarn test for stretchiness – these are all ways of sampling. They all help you test some aspect of your design.

Sampling is not a universal “should,” and a decision not to sample is certainly not a moral failure. Sometimes a sample is worthwhile, and sometimes it isn’t. Whether to sample is a judgment call that should be made on a project-by-project, sample-by-sample basis.

When should you sample?

Creating a sample is basically buying insurance. You do it to reduce the risk that you won’t like your finished design. Like insurance, the sample will cost you something – time, materials, and/or aggravation. And like insurance, the sample is only worth “buying” if the cost of the sample – both physically and emotionally – exceeds the likely cost of NOT doing the sample.

Now, how much does it cost you to skip the sample? It depends on two things: the probability of making a mistake, and the cost of the mistake if it does happen. 

If you’re working with familiar yarns, drafts, and colors, you can predict with reasonable confidence what’s going to result. The likelihood of a problem is relatively low.

If, on the other hand, you’re working with projects that involve new techniques or materials, the probability of a problem is considerably higher.

The next thing to consider is how much will it cost if you DO make a design mistake? Not just in money, but in time, materials, and frustration. A mistake that wastes $3 worth of yarn but which takes you 10 hours to correct, gritting your teeth the entire time, is still an expensive mistake to make!

Now, weigh the cost and the likelihood of error against the cost of your sample. Again, this is measured in time, money/materials, AND frustration. If you hate sampling, then the cost of the frustration should be worked into the “price”.

If the sample is less expensive than the likely cost of the error, then it’s worth buying the “insurance” and doing the sample. If it’s more expensive, then it isn’t. If the costs are close? Use your judgment.

If your risk is high and the stakes are high, and the sample is inexpensive, then it’s probably a good idea to make the sample. If, on the other hand, both the risk and stakes are low, the sample probably isn’t worthwhile (unless it’s super-easy to make).

How to choose a sampling method

As mentioned previously, there are many types of samples. Often more than one of them will work to test your design. For example, if you’d like to test your planned colors, you can do that via sketches, coloring squares on graph paper, computer simulations, card wraps, and/or woven samples.

To choose a method, pick the cheapest sample that will produce the answer you need.

This will vary depending on what stage the project is in. At the beginning, when the design is still pretty vague, a quick test – just enough to tell you you’re on the right track – might be just the thing. Towards the end, when you are finalizing your design, a more accurate (though more expensive) test might be in order.

There are lots of different ways to sample. You’ll find twelve of them – from sketches at the beginning to woven samples at the end – in our class Sampling: A New Approach. With each sampling method, you’ll get a description of the method, its advantages/disadvantages, and a summary of where it’s most effective. There’s even a “cheat sheet” that shows when to use each sample!