8 ms·
I finally read that book a few weeks ago, and one idea Graeber raises in it is something that'd been bugging me for a while: it seems like adding computers to a
by asark 7y ago
I finally read that book a few weeks ago, and one idea Graeber raises in it is something that'd been bugging me for a while: it seems like adding computers to a large proportion of activities and processes makes them less efficient, and increases the load of bullshit jobs to deal with making everything compatible with computer monitoring and computer-"aided" business processes, for gains that may be illusory and which surely do not make up for all the added layers (and cost) of bullshit.
The stuff about the rise of the influence of finance folks and practices on (previously) traditional management staff was interesting, too. I'm not sure his timeline works out, as I'm aware of (fictional) books that treated of precisely this sort of thing (bullshit office jobs, in particular) as an entrenched phenomenon at least as far back as 1961, but I think both ideas are getting at something worth further examination.
- ThrustVectoring 7y agoBureaucratic processes tend to get more and more complicated until the host organization is barely able to manage it. Computers allow better process management, so it allows complexity to expand for little overall gain.
- specialist 7y ago" adding computers to a large proportion of activities and processes makes them less efficient" Young technophilic me didn't distinguish between digitization and automation. I've since seen this mistake over and over. Automation is the removal of judgement. Automate before you digitize. Older technoskeptic me now assumes that whenever a big bang project fails (eg new payroll system for big org), it's because they tried to faithfully reimplement some legacy "processes" dependent on humans, in the form of social cognition and tribal knowledge, sometimes also called "culture". Also, automate after you digitize. In fact, redesign and reengineer to support automation.