I promised that I would post a round-up of the various components of this design. They have all appeared in this blog, but scattered over more than a year – a fairly realistic illustration of my scattered thought processes.
The overall design is a very personal one, incorporating three different ideas.
One is the use of statistical data as a source of pattern, specifically the Scottish Index of Multiple Deprivation which I described here. I have used the SIMD data to derive miniature portraits of various towns and villages. Well, I think of them as portraits, though in practice they are basically bar charts. Each block in the finished piece is a woven bar chart roughly 5″ long which portrays one place.
The second idea comes from old maps: the linear kind, such as maps for pilgrims, which show the desired journey as a more-or-less straight line and which highlight places to rest, stay overnight etc. I wanted to weave a linear map which would be autobiographical, highlighting places which have been important to me since I came to live in this area more than 20 years ago. So I sat down and made a list of 20 years’ worth of place-connections to be woven in chronological order.
The final component is the river. The line that links together all my places – indeed, the whole region – is the Tay. I decided to focus on the tidal stretch of the river, even though some of my places lie further upstream. This was mainly because it is the Firth of Tay I see every day. Its moods and rhythms are the moods and rhythms of daily life in Dundee. The Tay estuary is mainly shallow and the surface of the water is very lively: lots of small choppy waves all bumping into one another. This is why I chopped up the blocks into segements where the colours are reversed.
That’s a brief summary of the motivation; now a brief summary of the practical side.
I have 16 shafts at my disposal and I wanted to use double weave with 8 shafts for each layer and a gradual layer exchange (switching one pair of shafts at a time). My sampling began with this warp threaded for an advancing twill on 8 shafts in each layer – but with a completely different plan in mind. I loved some of the effects but this is the wrong idea for them: they need to be developed with an idea of their own.
The next sample warp had a simple straight draw threading, which was a bit dull but allowed me to try the liftplan I’d previously worked out and see how it scaled to my proposed weight of warp yarn: two strands of 60/2 silk per end. I found that even though the layer changes are frequent, the fabric was still quite slippery and it benefited from the introduction of a stitch here and there. (I referred to the excellent article by Doramay Keasbey in The Magic of Doubleweave to work out what I needed.)
Being happy with the weight of yarn, I then sampled for colour. The Tay is famously silvery, of course, but I found that the silver-coloured silk on its own was a bit too pink: OK for some highlights, but I wanted more blue. The threading I used here is the one I settled on. It’s a more rapid advance than the one in the first warp so the blockiness is less pronounced, but it’s still choppier than the second version.
I tried a whole host of wefts on the last sample warp and liked them all, so a final choice was very difficult. I wanted to have a single unifying weft not lots more choppiness, and settled on a mid-value grey bourette silk which gives the finished fabric a satisfyingly rough texture.
The last step was to test the whole plan with a mock-up in Excel. I couldn’t model this in Fiberworks without spending hours copying and pasting specific units of the liftplan. (My chain needed to be 72 lags, eight for each of nine possible layer combinations, and the maximum Fiberworks tie-up is for 64 treadles.) But it is amazing what a lookup function combined with conditional formatting can do.
The ‘extreme’ blocks at the right-hand end of the chart represent one particular place which has quite a skewed SIMD profile. I wasn’t sure whether it was too extreme for the rest, but decided to stick to the plan anyway. In fact, I like the extremes in the finished piece, maybe even more than the middling places, though it did make it difficult to take a ‘representative’ photo.