[-empyre-] Not digital! Analog computing and PRACTICE

[Quinn DuPont]

It’s late in the day, and late in the week, but there are still a few threads intriguing me. Most especially, the “opposite” of this month’s theme: analog.

The discussion of analog vs./and/or digital discussed by everyone, but especially Ange Albertini, Andres Ramirez Gaviria, and John Hopkins, really exposes the intricacies and thorny issues present. At times the distinction becomes a purely technical one based on sampling, or a social one that responds to — and resists — production, or even a personal one invested in “creativity” or “innovation”. I’m reminded of Wendy Chun’s (“Programmed Visions”, p. 146 ff) mathematical and technical description of sophisticated early analog computers, especially the differential analyzer. She writes, “intriguingly, direct representation — or more accurately correspondence — makes analog machines live, vivid, and direct.” In contrast, digital computers, “by hiding mimesis, could simulate any other machines… [but,] the [digital] computer becomes a simulacrum, rather than a simulation.” I find Chun’s perspective so refreshing because it makes a case for analog computing not just because they are faster (they are, depending on how you “count”), but simply because they are more *real*. The downside of digital objects is precisely that they aren’t really for humans, or, perhaps, even about human things. Ashley’s Scarlett’s worry that these processes occur outside of human perception, and the discussions of PRACTICE throughout the week, seem to be entirely cognizant of this deficiency. Perhaps we need to think more seriously about analog computing!
(My favourite analog computing apparatus is spaghetti, which makes a mighty efficient sort algorithm: https://en.wikipedia.org/wiki/Spaghetti_sort)

And, speaking of the sub-perceptual matters of PRACTICE, Ange Albertini’s discussion of the non-digital of digital apparatuses reminded me of Kryoflux (http://www.kryoflux.com/), a tool used by digital archivists to forensically preserve floppy disks. Perhaps Ange is familiar with the tool, since he even mentioned how fuzzy bits were frequently used for copy protection. Indeed, the schemes used to physically instantiated copy protection in floppy disks were inventive! Long sectors, timing (in)precisions, and strange write methods were used to ensure that disks could not be copied, or if copied, could be detected by software. The advantage of a tool like Kryoflux is that when copying digital data it does not copy at the level of the digital, instead, it copies magnetic flux changes. Matt Kirschenbaum uses Kryoflux when performing digital forensics at the Maryland Institute for Technology in the Humanities, and has written a little about their setup: http://www.bitcurator.net/2013/08/02/building-a-digital-curation-workstation-with-bitcurator-update/

Unfortunately, the theoretical underpinnings of Kryoflux are still somewhat under-explored (again, Kirschenbaum is the point person here, with his “Mechanisms”). The cross-over between PRACTICE, analog, and digital is riveting. I was once working on a paper at these intersections using Kryoflux as a central lever, but it never got anywhere, so, in its stead, here’s an image of Kryoflux’s software analysing a disk. 
http://www.softpres.org/_media/kryoflux:ui:kfui-scat-bad.png?w=&h=&cache=cache
The green blocks are good sectors of the disk, the red a bad sector. It has detected the disk format (and thus can sort out sector data), in this case it is an Apple DOS disk. And finally, the plot on the right show the actual magnet flux. This plot of the magnetic flux isn’t just a pretty picture, with a trained eye one can actually detect abnormalities such as copy protection schemes.


~Quinn DuPont (iqdupont.com)