But now that I am into those charts by several rows, I'm finding a part of the design that really bugs me.
Two yarnovers that I put between each of the point in the border don't really blend properly with the rest of the yarnovers that outline the points. (Interestingly, they are also the cause of my miscount in stitches from the border to the body of the shawl.) If they were gone, then the outline would blend almost seamlessly from one point to the next. I didn't notice this on my swatch, possibly because the yarn was lighter and possibly because I didn't work up the body of the shawl in the swatch other than a row or two to cast off. It is one of those examples of something that looks right on the charts, however, because of the way the yarnovers interact with other knit stitches and the decreases, it looks wrong in real yarn.
My best choice for making a sample that looks right is to frog the whole thing and start over. The change affects even the amount of stitches in the cast on, so I really would have to start from scratch. I could keep going and revise the final pattern for the test knitters only, but then I wouldn't have a good sample to photograph.
It's a painful thought, but I have worked with the philosophy with my handwork lately that it's better to stop and fix things when they are wrong rather than assume they will work out later. The latter seems to leave me with projects that I don't enjoy and won't use. This shawl has been in process since September when I first tried to use part of the design and found it didn't work out as I expected.
While I'm at it, I may even change yarns. The color of this one hasn't thrilled me as much as I hoped. Of course, that would mean more swatching....