When Computers Were Women (work in progress)

This is a project I’ve been working on since late 2018 which is still in progress due to production trips being postponed because of COVID. The work is connected to my practice-based PhD research on feminist critiques of physics and stems from an invited residency at CERN in 2018.

Following that visit, I pitched an artwork—a film about the non-scientific labour (cooking/cleaning/housekeeping/administration/grounds maintenance) mostly performed by women which effectively enables physicists to carry out their own work in a relatively frictionless environment. Permission wasn’t granted and that was basically the end of that.

But during that visit, I spotted a handful of computer programming punchcards in a cabinet in the data centre and was struck by their resemblance to a much older form of data-processing technology: the long, thin punch cards used to control the rods and hooks that raise the warp threads of looms fitted with Jacquard devices.

There are any number of tedious tech articles on the relationship between Jacquard’s punch card-based mechanism and Charles Babbage’s Analytical Engine that proclaim Jacquard’s genius and embrace him as inventor of the first proto-computer. That said, when I happened upon the cards at CERN, the connections between histories of computational and weaving technology were too generative to ignore. Not only were there parallels between sexism in physics and technology, namely, the erasure of women and people of colour from conventional histories that tend to glorify white male figures. But there were also correspondences between the use of gendered, sometimes ‘low skill’ labour—women in weaving as pattern readers; women in physics as “computers” or “scanner girls”—and the later evolution of such gendered work into labour-saving computerised algorithms.

I wondered whether there were more of the punch cards at CERN and returned for a second visit early this year. Excellent sleuthing by the data centre’s communications director and a physicists from the ATLAS experiment had turned up an entire programme perfectly preserved in a basement storage room. During the second visit, I individually photographed the programme’s 2,131 punch cards. Thanks to information provided by curious engineers who occasionally stopped by to see what I was doing, I also managed to identity the original programmer (who had recently retired) and obtain information about the programme and its related experiment.

In addition to questions of gendered labour, labour-saving technologies and computational history, the project also looks to translation—how to move between different forms and fields. You might think it’s an easy thing to translate computer punch cards into Jacquard punch cards, but it isn’t.

Luckily, I was saved by the Python know-how of Henry Cooke (aka @prehensile), who helped me enormously by reconfiguring an existing script that scanned the punch cards I’d photographed, extracted their binary data and reconstructed everything in the form I needed to create the weaving grid.

Here’s an image of the final grid created by reading the data of the CERN punch cards and translating it into a form that can (theoretically) be read by a Jacquard loom mechanism—basically the set of instructions that tells the loom what to do.

Once I can travel again, the final stage of the process will be to actually weave the tapestry (actually, a series of tapestries). What’s fascinating is that I have no idea what the finished piece will look like as the weaving grid doesn’t represent a design, but a set of instructions for the loom to follow.

Further Projects