I love the term "software archaeology" because it implies the existence of subfields such as "software experimental archaeology" (attempting to reconstruct and demonstrate how people once built software) and "software paleoethnobotany" (quantifying what botanicals were culturally significant to software and the broader historic implications to the societies that wrote it), but also the existence of the broader field "software anthropology" and its offshoot "software sociology", and
@aeva As a person who has done a lot of software archeology in past jobs, I can definitively tell you that "software pseudoarcheology" is also alive and well in organizations.
@aeva In my current project, I often use the term Git Archaeology. Which I think applies here. - "git experimental archaeology" What problem was this commit intended to solve? - "git anthropology" What was the culture and pressures that created these design decisions? - "git paleoethnobotany" WTF WERE THEY ON!?
You might say to yourself "Aeva, there's no such thing as Software Paleoethnobotany that's absurd", but some day a thousand years from now someone is going to be tasked with analyzing the remains of a surprisingly well preserved set of backup disks for an ancient industrial automation system to reconstruct the lost recipe for brewing the legendary "miller high life"
@aeva Published November 3057: Relative efficacy of indoor hydroponic software focused on cultivation of Sativa strains.
Abstract We study a number of intact hydroponic devices that date to an era immediately before the second North American dust bowl. Comparative analysis of the installed software strongly suggest that this is not an antecedent of the common Strawbery2 hydro system which provided homesteads with most of their calories during the early 2250s.
@aeva I do entirely too much config archeology at work when we try to work out why the hell a specific setting was made over a decade ago. Sometimes, we have a useful change ticket… but most of the time? It just says “change X to be Y” ?
@aeva The repository handling scripts in Arch Linux is around 20 years old now. Rewritten twice over three different VCS tools.
It's also been developed in cvs, SVN and git, and the all the history is preserved!
I have several times run the git log looking at the 20 year old code wondering how they did stuff way back when to make sense of current repo handling.
@aeva this fits with a term I learned when I joined a startup. I asked about "documentation" to get up to speed, and I was told, it was "tribal knowledge". Understanding the code was passed on from Developer to Developer and flowers down from the "elders".
"software cryptozoology": a pseudoscience and subculture that searches for and studies the benefits of blockchain technologies whose present existence is disputed or unsubstantiated
@Stefan_S_from_H@aeva I love A Fire Upon the Deep too, one of my favorites ever, but the software archaeology where they discovered the Unix epoch was in en.wikipedia.org/wiki/A_Deepne… , which is still on my reading list.
@aeva I've been thinking for about five years about teaching software as anthropology, in fact. It's all about people and behavior, whether users, peers, management, or predecessors.
@aeva '"software experimental archaeology" (attempting to reconstruct and demonstrate how people once built software)' Yeah... did that last year... Had to fix the bootloader for one of our legacy cameras that hadn't been touched in 10 years. Toolchain: I found multiple pieces scattered on servers and backups of former colleagues' computers. Sourcecode: Multiple source-tree copies + Git-conversions of the SourceSafe copies... None of which produced the exact bootloader that we used in production... In the end I reverse engineered the toolchain and the sourcecode to the point that I got a binary identical binary (minus timestamps). :blobcatuwu: