Veronika Irvine is a postdoctoral fellow in the Cheriton School of Computer Science at the University of Waterloo, Canada. She studied math and computer science and received her Masters degree from the University of Waterloo and her Ph.D. from the University of Victoria. Her research area is the computational generation of textile structures based on the technique of bobbin lace.
Sigrid Adriaenssens: What is bobbin lace?
Veronika Irvine: Bobbin lace is a 500-year-old fibre-art form in which fine threads are braided together to form delicate patterns. You may have encountered it in the elaborate ruff collars of the Elizabethan era or the large square collars in paintings by Vermeer and Rembrandt .
I like to compare the process of making bobbin lace to the more familiar process of weaving on a loom. Both processes can use hundreds of threads in a single piece. In weaving, all but one of the threads (called the warp) are held in place by a scaffold called a loom. The remaining thread (called the weft) is wrapped around a shuttle and weaves back and forth through the warp. In bobbin lace, the threads are also interlaced. The main difference is that in bobbin lace the threads are not restricted to just horizontal or just vertical directions. Each thread is wrapped around a bobbin much like the weft thread. It can travel horizontally, vertically, at an angle and can switch direction at any time. This increased flexibility in direction comes with a restriction: the interweaving of threads must strictly alternate between over and under to lock the threads in place. In loom weaving the over and under patterns is much more varied giving rise to fabrics such as twill in which the weft thread goes over one or more warp threads then under two or more warp threads with an offset between rows. Another key difference is that in bobbin lace there is no loom. Instead the lacemaker constructs scaffolding as the piece progresses, placing pins at any point where a thread needs to change direction or to resist the tensioning forces she applies.
Bobbin lace pieces typically have woven motifs (plain weave, triaxial weave or a mix of both to make motifs such as flowers, geometric shapes, animal and human figures) intermixed with more complex repeated patterns which can form a net in the background or a more decorative filling to enhance the motif. Lace is traditionally made from a single colour of thread but it is capable of expressing a wide range of shades through the diverse density of threads.
SA: What are the phenomena/rules found in bobbin lace that attract you as a computer scientist?
VI: In spite of its seemingly complex appearance, designing bobbin lace has many problems that can be stated fairly simply. However, the tools required to solve these problems can be complex or even unknown. The most challenging aspect I have encountered is the `conservation of threads’. In a continuous rectangular piece of bobbin lace, the entire piece from top to bottom is made with one set of threads. You cannot add or remove threads. So when a thread is required to make a braid at a particular point in the design, you must plan the path of the thread from the top border of the fabric to that point, so that it will be available. In my Ph.D. thesis, I expressed this requirement as a set of rules governing the network of threads that make up a pattern.
It is easy to test whether a particular network satisfies these rules and this can be done in a reasonable amount of time by a computer. The interesting part is to find families of networks or efficient algorithms for generating new networks that will pass this test. I look for inspiration in places such as tiling theory and Islamic ornaments. Then I do background research into the mathematics behind the subject in order to understand the relationships between the various parts and whether these relationships will enforce the rules of bobbin lace. It has led me to learn about many different areas of mathematics such as topology, group theory, knot theory and graph theory.
SA: What can we learn from those rules?
VI: One of the key observations from my research is that there exist an infinite number of these repeated filling patterns. Traditional bobbin lace has many regional styles, each defined in a large part by the set of fillings used. For example, Chantilly lace is a particular style of bobbin lace practiced originally in the Bayeux region of France and it uses filling patterns such as fond simple, point de Paris and fond de mariage. Today bobbin lace is often thought of as ‘antique lace’, a traditional art form that has fully matured and belongs to a previous era. However, discovering that there are an unlimited number of fillings means that this art form is still in its infancy, there is much yet to explore. In addition to the artistic exploration of new styles in tune with our modern aesthetic, this construction technique can produce fabrics with material properties not found in knitting or weaving, textiles with applications in architecture, medicine, sports equipment or space exploration. I am very interested in finding out more about the strength of these fabrics, what properties are possible and developing algorithms to generate new networks that will realise these properties.
SA: How does your computer science research influence your lace making? Can you give some examples?
VI: Understanding the logic behind these complex networks has freed me to design my own pieces. Recreating pieces of the past has many challenges but for me the thrill of creating something that has not been done before is very exciting. So in my lace practice, I am always looking for some new idea. I do this by pushing the rules to their limits and in some cases exploring what happens when you break them. For example, traditionally these filling patterns have always been periodic. Recently I have been studying quasiperiodic patterns. These are patterns, such as the Penrose tiling by thick and thin rhombs, that have regular repetition of elements but which cannot be created by translating a fundamental region. With Craig Kaplan and Therese Biedl, I was able to prove the existence of three families of quasieriodic lace patterns and conjecture about the existence of others. The pieces “Nodding bur-marigold” and “Ammann’s web” represent two of these families.
SA: Which of your bobbin lace creations are you most proud of and why?
I don’t really have a favourite. It is exciting and interesting to explore new possibilities and whatever project I am currently working on is the one that has my full attention. Recently I have started to explore sculptural forms and I really enjoy the interplay of form with shadow and light. The transparent property of lace brings a level of richness to a surface in three dimensions as it twists and wraps around itself, creating layers and new patterns through overlap.