Network patterns in paisley

Researching the history of the internet I came across this image in Wiki Commons. It is a diagram of the NSFNET backbone on paisley notepaper, dated 1 October 1990, by B.S. Blair (CC BY-SA 4.0). I am fascinated by this palimpsest so will expand on my original social media post below.

In July 1986 the National Science Foundation Network (NSFNET) backbone became fully operational, linking the National Center for Atmospheric Research (NCAR) in Boulder, Colorado to five research centres: Princeton University (Princeton, NJ), Cornell University (Ithaca, NY), the University of California San Diego (UCSD), the University of Illinois Urbana-Champaign (UIUC), and the University of Pittsburgh. This diagram shows several new nodes: Seattle, Palo Alto, Salt Lake City, Lincoln, Houston, Atlanta, University of Maryland (College Park), and Ann Arbor (Merit Network). 

This diagram represents a noble plan to link research and education through telepresence, though the infrastructure had been borrowed from the military complex. As is well known, the predecessor to this system had been built as the Advanced Research Projects Agency Network (ARPANET) by the US Department of Defense, starting 1969.

Here each dot represents the ultimate in computing power at the time, as expressed by bandwidth. Each line represents an abstraction of cables laid under the soil. The computer centres and pathways each required a significant investment, tabulated in balance sheets and budget reports. Mentioned less often, if at all, were the workers who built this engineering marvel. Even computer programmers (e.g. "calculators") were largely uncredited by name. This activity was also seen as a form of manual labour not different from the typing pool.

The other significant symbol in this image is the boteh, an ornamental teardrop shape from Persia. This might represent the sun, a phoenix, or simply a leaf with its patterns of circulation emphasised. It’s idealised like the network diagram but curvilinear. As it became increasingly popular in Western Europe in the 19th century the boteh pattern in fabric was renamed Paisley after the Scottish town. This place had been known for two centuries as a centre for textile production.

As documented in “The Paisley Shawl and the People Who Made It” (1904) by Matthew Blair (available online), Paisley workers worked manually in the main. Although the Jacquard loom was invented in 1804, it was not until the mid-century that this automation replaced the work of the draw-boy in Paisley.

The Jacquard head used a punch-card system that encoded the pattern for the loom. Herman Hollerith expanded on this idea for an Electric Tabulating System first used to compile data from the 1890 United States census. His Tabulating Company invented the keypunch machine and automated card feeding, preparing the ground for 20th century computer data entry. When I first learned FORTRAN coding in the late 1970s, every program was a stack of punch cards. My typing lessons inculcated skills useful for both calculations and linguistic output.

The abstract linearity of the telecommunications network and the naturalistic curves of the floral patterns form a dialectic that obscures such hidden connections. But there is something more here.

B.S. Blair (whoever he was) did not "draw" this diagram. The numeric codes and locations are clearly the result of some sort of mechanistic impression, not a pen or pencil. Given the technology of the time, it's likely plotter or line printer output. So why then the paisley?

Indeed, by appearances the floral pattern has been drawn on the engineering diagram, not the other way around. If readers have insights, I would love to hear them.