Tuesday, August 14, 2007

Toward a Diagnostic Instrument for Pre-Design Chromatic Evaluation

Which is to say, I think I've come up with a way of talking with people about the color schemes they would like in a quilt that's being made for them.

Statement of Problem: Why bother? Well, I've found that people (including me) have a hard time talking about color. Everyone has their favorite colors, of course, but "green" is not really enough to design a quilt around. People who are color savvy will sometimes list their favorite color combinations, and a rare individual will also throw in an idea of color quality, with words like "pastel" or "jewel tones" or "smoky."

But since a quilt design needs to encompass not just color, color combinations, and color quality, but also color value and the relative proportions of all of the above across the quilt surface, it's pretty brutal to talk about without looking at quilt images. So, you end up flipping through a stack of quilt books and magazines, looking at a random selection of posibilities, slapping down sticky notes everywhere, and it's a huge mess. Well, no. It's not really that huge of a mess. I'm just trying to get you all excited about my innovation.

Innovation: I selected 38 images that represent a wide range of possible quilt color combinations, and mounted them on cardstock, like so:

Methodology: The subject is instructed to sort the cards into three piles based solely on colors. One pile is the color combinations that the subject really likes, the second is the color combinations that are "all right," and the third is the color combinations that the subject doesn't like. When the piles are complete, the interviewer and subject examine the stack together, looking for common themes in the subject's choices.

Results: I tried this out with Mrs. 5000, who is always ready for anything, at lunch today. It worked, in the sense that I learned things about her color taste that I hadn't realized before. She is not crazy about extensive white or very light areas on quilts, for instance, and has more of a thing for jewel tones than I expected.

Conclusions: Encouraged by this first trial, I will apply the procedure to Neice #1 this coming weekend, in preparation for beginning the design of her graduation quilt. Hopefully, it will help me come up with a quilt that will be all, like, aesthetically pleasing to her.


jovaliquilts said...

So clever -- love the concept! Will be interested to see if it works. If so, will you market the cards? I find it hard to have "help" from the people I'm making quilts for. It's difficult for non-quilters to see a block or pattern and separate it from the particular fabrics in it, and it's hard to see a bolt of fabric and understand how it will work, in proportion, with the other fabrics and the overall design.

Libby said...

Yup pitch it to Fons & Porter and make your first million :-) It's a really good idea...as much as I love attempting to guess what I think my quilt recipients will like, it seems sooo much smarter to do it this way - or any other way.

Rebel said...

You realize, you could also just send the person down to the Fred Meyer pain department and ask them to pick out a couple color cards that they like. 'Course your way is a lot cooler. ;)

Mrs. 5000 said...

It was quite fun to be the guinea pig evaluatee. I feel like Michael knows my aesthetic preferences pretty well, so the fact that he learned something new speaks well for the evaluation process. Probably to be marketable, you'd want it to have a pretty multi-layered analysis. A little more complicated, with lots of pretty charts. Or else a personality type diagnosis: "Are you a Pastel? I'm a Jewel Tone. You value family and self-sacrifice; I'm imaginative and somewhat eccentric. How 'bout that!"

Michael5000 said...

They worked! I think!

Market them, though? I dunno. It would be an AWFULLY niche product, and you'd have to either make a million quilts or secure rights to use images of them, and that sounds like an enormous pain. Also, as (the always brilliant) Mrs. 5000 says, you'd probably want to throw in some cutesy mumbo-jumbo to make 'em sell, and that would drive me bonkers.

@rebel: Well yeah, but the reason I made the cards is because using something like paint chips doesn't have enough quilt context to it. One things I'm noticing from the cards is that people tend to group them according to overall intensities of color or levels of value contrast, rather than according to colors per se. It's interesting, and not something you'd find out from paint chips.

A little story: on the way to Sisters, I decided I was going to try to find (and announce on this blog) my favorite quilt for each of the seven traditional spectral colors, plus grey and brown. Once I got to the show, I realized immediately that it was a completely stupid idea, and was embarassed that I had thought of it after 13 years of quilting. Why? Because quilts are rarely in a single range of color. Well under 20% in the Sisters show, by my count, could be said to BE a certain color.

I thought that was interesting. And then, when it came time to think about Niece #1's quilt, I thunk to myself, "I know her favorite color is green -- but to know this is to know nothing. I need to get her talking about some real quilts."

Wow, what a ridiculously long comment I just made.