I have been doing swatches to illustrated my index of knitting stitches for almost two years. Up until the first 1000 stitch patterns had been illustrated, I admit, the selection of a stitch to work was biased toward the easy patterns just to pump up the numbers.
Now that I have a passed the 1000 milestone and now that I have a firm grasp on the ‘50% of documented patterns are illustrated’ milestone, I have adopted a different attitude toward picking patterns to swatch. I want to focus on working the the biggest, most difficult stitches first. And being the database wonk I created a database view to help me. The algorithm is very simple: list the stitches in descending order by the product of the (number of stitches) x (number of rows). One side effect of this is that the number of illustrated stitches does not increase as fast. The other side effect is that more ‘high quality’ stitches are swatched.